import numpy as np from statsmodels.tools.testing import ParamsTableTestBunch est = dict( deviance=18.59164098607571, dispers=1.859164098607571, deviance_s=18.59164098607571, dispers_s=1.859164098607571, deviance_p=24.75374834715614, dispers_p=2.475374834715614, deviance_ps=24.75374834715614, dispers_ps=2.475374834715614, bic=-9.740492454486454, nbml=0, N=17, ic=3, k=7, k_eq=1, k_dv=1, converged=1, k_autoCns=0, ll=-31.92732830809848, chi2=128.8021169250575, p=2.29729497374e-25, rc=0, aic=4.579685683305704, rank=7, canonical=1, power=0, df_m=6, df=10, vf=1, phi=1, k_eq_model=0, properties="b V", depvar="executions", which="max", technique="nr", singularHmethod="m-marquardt", ml_method="e2", crittype="log likelihood", user="glim_lf", title="Generalized linear models", opt="moptimize", chi2type="Wald", link="glim_l03", varfunc="glim_v3", m="1", a="1", oim="oim", opt1="ML", varfuncf="u", varfunct="Poisson", linkf="ln(u)", linkt="Log", vce="oim", vcetype="OIM", hac_lag="15", marginsok="default", marginsnotok="stdp Anscombe Cooksd Deviance Hat Likelihood Pearson Response Score Working ADJusted STAndardized STUdentized MODified", # noqa:E501 predict="glim_p", cmd="glm", cmdline="glm executions income perpoverty perblack LN_VC100k96 south degree, family(poisson)", # noqa:E501 ) params_table = np.array([ .00026110166569, .00005187148786, 5.0336259178483, 4.812884279e-07, .00015943541766, .00036276791372, np.nan, 1.9599639845401, 0, .07781804809828, .07940260798777, .98004398180811, .32706440886796, -.0778082038363, .23344430003287, np.nan, 1.9599639845401, 0, -.09493110013466, .02291930335216, -4.1419714498302, .00003443332141, -.13985210925565, -.05001009101367, np.nan, 1.9599639845401, 0, .29693462055586, .43751760764129, .67868038993144, .49734039404176, -.5605841330232, 1.1544533741349, np.nan, 1.9599639845401, 0, 2.3011832004524, .42838381728481, 5.3717790159251, 7.796361708e-08, 1.4615663470144, 3.1408000538904, np.nan, 1.9599639845401, 0, -18.722067603077, 4.2839791307242, -4.3702518223781, .00001241033322, -27.118512409818, -10.325622796337, np.nan, 1.9599639845401, 0, -6.8014789919532, 4.146873025502, -1.6401464308471, .10097472438129, -14.929200770398, 1.3262427864914, np.nan, 1.9599639845401, 0]).reshape(7, 9) params_table_colnames = 'b se z pvalue ll ul df crit eform'.split() params_table_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov = np.array([ 2.690651253e-09, 1.942168909e-06, 9.445812833e-08, 4.703695025e-06, -6.082922480e-06, -.00008108248895, -.00013492774575, 1.942168909e-06, .00630477415526, .00017467012687, .00328093520848, -.01768604570302, .11117887243846, -.19441636422025, 9.445812833e-08, .00017467012687, .00052529446615, -.00313545508833, -.00516707569472, -.03253594627601, .01688876616272, 4.703695025e-06, .00328093520848, -.00313545508833, .19142165699616, -.00179497953339, .30391667530759, -1.4489146451821, -6.082922480e-06, -.01768604570302, -.00516707569472, -.00179497953339, .18351269491151, .3016848477378, .36484063612427, -.00008108248895, .11117887243846, -.03253594627601, .30391667530759, .3016848477378, 18.352477192481, -4.0741043266703, -.00013492774575, -.19441636422025, .01688876616272, -1.4489146451821, .36484063612427, -4.0741043266703, 17.196555889636]).reshape(7, 7) cov_colnames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] infocrit = np.array([ 17, np.nan, -31.927328308098, 7, 77.854656616197, 83.68715002459]) infocrit_colnames = 'N ll0 ll df AIC BIC'.split() infocrit_rownames = '.'.split() predicted = np.array([ 35.226364135742, .16681243479252, .98022246360779, 8.1965742111206, .33106967806816, .89840310811996, 1.3118965625763, .29945519566536, .11764223873615, 3.6862981319427, .35516858100891, .46500706672668, 2.0823004245758, .3434439599514, .24561515450478, 1.0650315284729, .62310123443604, .41350400447845, 1.9260421991348, .40797635912895, .32057955861092, 2.4171404838562, .36215576529503, .31702440977097, 1.8473218679428, .3869916498661, .27665960788727, 2.8643238544464, .43869277834892, .55124300718307, 3.1211984157562, .44224792718887, .61045408248901, 3.338207244873, .42789322137833, .61120104789734, 2.5269968509674, .42458593845367, .45554983615875, .89725440740585, .59187793731689, .31432569026947, .97933322191238, .37813624739647, .14003194868565, .53462094068527, .38791963458061, .08045063912868, 1.9790935516357, .31954729557037, .20208616554737]).reshape(17, 3) predicted_colnames = 'predict_mu predict_linpred_std predict_hat'.split() predicted_rownames = ['r'+str(n) for n in range(1, 18)] resids = np.array([ 1.773634314537, 1.773634314537, .29638093709946, .29637759923935, .2988341152668, .05034962296486, .80342543125153, .80342543125153, .27623143792152, .27622014284134, .28062695264816, .09801965206861, 4.6881031990051, 4.6881031990051, 3.0157172679901, 2.977787733078, 4.0930528640747, 3.5735311508179, .31370183825493, .31370183825493, .1611547768116, .16114975512028, .16338862478733, .08509942144156, .91769951581955, .91769951581955, .59656941890717, .59618371725082, .63595855236053, .44071426987648, .9349684715271, .9349684715271, .80822360515594, .80661898851395, .90597397089005, .87787866592407, .07395775616169, .07395775616169, .05295527353883, .05295492336154, .05329062789679, .03839882463217, -.41714036464691, -.41714036464691, -.27668312191963, -.27663832902908, -.2683065533638, -.17257598042488, -.84732186794281, -.84732186794281, -.68459099531174, -.68349820375443, -.6234148144722, -.458675801754, -1.8643238544464, -1.8643238544464, -1.2799508571625, -1.274356007576, -1.1015654802322, -.65087747573853, -2.1211984157562, -2.1211984157562, -1.4092296361923, -1.4021278619766, -1.2006615400314, -.67961025238037, -2.338207244873, -2.338207244873, -1.5136297941208, -1.5051733255386, -1.2797535657883, -.70043802261353, -1.5269968509674, -1.5269968509674, -1.0992211103439, -1.0954134464264, -.9605849981308, -.60427337884903, .10274560004473, .10274560004473, .10649761557579, .1064917370677, .10846894979477, .11451110988855, .02066676132381, .02066676132381, .02081091701984, .02081087417901, .02088368684053, .02110289037228, .46537905931473, .46537905931473, .56824368238449, .56713002920151, .63647866249084, .87048417329788, -.97909361124039, -.97909361124039, -.77151334285736, -.77000600099564, -.69597083330154, -.49471819400787]).reshape(17, 6) resids_colnames = ['score_factor', 'resid_response', 'resid_anscombe', 'resid_deviance', 'resid_pearson', 'resid_working'] resids_rownames = ['r'+str(n) for n in range(1, 18)] results_poisson_none_nonrobust = ParamsTableTestBunch( params_table=params_table, params_table_colnames=params_table_colnames, params_table_rownames=params_table_rownames, cov=cov, cov_colnames=cov_colnames, cov_rownames=cov_rownames, infocrit=infocrit, infocrit_colnames=infocrit_colnames, infocrit_rownames=infocrit_rownames, predicted=predicted, predicted_colnames=predicted_colnames, predicted_rownames=predicted_rownames, resids=resids, resids_colnames=resids_colnames, resids_rownames=resids_rownames, **est ) est = dict( deviance=23.34969514421719, dispers=.8980651978545075, deviance_s=23.34969514421719, dispers_s=.8980651978545075, deviance_p=30.06164170990202, dispers_p=1.156216988842385, deviance_ps=30.06164170990202, dispers_ps=1.156216988842385, bic=-67.5595014539113, nbml=0, N=33, ic=3, k=7, k_eq=1, k_dv=1, converged=1, k_autoCns=0, ll=-52.96941847346162, chi2=183.6836771894393, p=5.59891844113e-37, rc=0, aic=3.634510210512826, rank=7, canonical=1, power=0, df_m=6, df=26, vf=1, phi=1, k_eq_model=0, properties="b V", depvar="executions", which="max", technique="nr", singularHmethod="m-marquardt", ml_method="e2", crittype="log likelihood", user="glim_lf", title="Generalized linear models", opt="moptimize", chi2type="Wald", wtype="fweight", wexp="= fweight", link="glim_l03", varfunc="glim_v3", m="1", a="1", oim="oim", opt1="ML", varfuncf="u", varfunct="Poisson", linkf="ln(u)", linkt="Log", vce="oim", vcetype="OIM", hac_lag="15", marginsok="default", marginsnotok="stdp Anscombe Cooksd Deviance Hat Likelihood Pearson Response Score Working ADJusted STAndardized STUdentized MODified", # noqa:E501 predict="glim_p", cmd="glm", cmdline="glm executions income perpoverty perblack LN_VC100k96 south degree [fweight=fweight], family(poisson)", # noqa:E501 ) params_table = np.array([ .00025343868829, .00004015414514, 6.3116444744157, 2.760858933e-10, .00017473800999, .00033213936659, np.nan, 1.9599639845401, 0, .09081422305585, .06472607217881, 1.4030547505642, .16060051303473, -.03604654727537, .21767499338706, np.nan, 1.9599639845401, 0, -.09416451429381, .01795769655821, -5.2436855689475, 1.574003474e-07, -.12936095279319, -.05896807579442, np.nan, 1.9599639845401, 0, .27652273809506, .38626128010796, .7158955669017, .47405583598111, -.48053545953887, 1.033580935729, np.nan, 1.9599639845401, 0, 2.239890838384, .36339399714255, 6.1638080320445, 7.101602988e-10, 1.5276516917866, 2.9521299849815, np.nan, 1.9599639845401, 0, -18.842583191417, 3.736940161486, -5.0422491067996, 4.600917913e-07, -26.16685132031, -11.518315062523, np.nan, 1.9599639845401, 0, -6.5630017977416, 3.2352486362722, -2.0285927097411, .04249979172538, -12.903972605867, -.22203098961573, np.nan, 1.9599639845401, 0]).reshape(7, 9) params_table_colnames = 'b se z pvalue ll ul df crit eform'.split() params_table_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov = np.array([ 1.612355372e-09, 1.270985149e-06, 8.789752394e-08, -1.636449642e-07, -3.213686689e-06, -.00005643188411, -.00006199883309, 1.270985149e-06, .0041894644197, .00016567874308, -.00066453618021, -.00943379587945, .07218307550995, -.11262571631082, 8.789752394e-08, .00016567874308, .00032247886568, -.00355795369216, -.00391377556228, -.01880905186772, .01900717143416, -1.636449642e-07, -.00066453618021, -.00355795369216, .14919777651064, .02481983169552, .26952997380446, -.95915288407306, -3.213686689e-06, -.00943379587945, -.00391377556228, .02481983169552, .13205519715924, .44364186152042, -.0298149336078, -.00005643188411, .07218307550995, -.01880905186772, .26952997380446, .44364186152042, 13.964721770527, -3.6510403528048, -.00006199883309, -.11262571631082, .01900717143416, -.95915288407306, -.0298149336078, -3.6510403528048, 10.466833738501]).reshape(7, 7) cov_colnames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] infocrit = np.array([ 33, np.nan, -52.969418473462, 7, 119.93883694692, 130.41438987719]) infocrit_colnames = 'N ll0 ll df AIC BIC'.split() infocrit_rownames = '.'.split() predicted = np.array([ 34.815238952637, .16658315062523, .96612107753754, 7.3026847839355, .32757967710495, .78363972902298, 1.2540435791016, .26076200604439, .08527097851038, 3.9734709262848, .24942673742771, .24720433354378, 2.0739872455597, .24682784080505, .12635557353497, 1.1471545696259, .45427960157394, .23673823475838, 1.7763512134552, .27608770132065, .13540133833885, 2.2698366641998, .25641229748726, .1492355465889, 1.6349502801895, .27634221315384, .12485299259424, 2.7504913806915, .39550569653511, .43024495244026, 2.862185716629, .39729079604149, .45176732540131, 3.5617923736572, .39150056242943, .54592549800873, 2.6135795116425, .29556328058243, .22831618785858, .775799036026, .40655690431595, .12823067605495, .93375068902969, .29390665888786, .08065843582153, .56681954860687, .28863781690598, .04722274839878, 1.8914022445679, .21889741718769, .09062857925892]).reshape(17, 3) predicted_colnames = 'predict_mu predict_linpred_std predict_hat'.split() predicted_rownames = ['r'+str(n) for n in range(1, 18)] resids = np.array([ 2.1847612857819, 2.1847612857819, .36650228500366, .36649596691132, .3702706694603, .06275302171707, 1.6973150968552, 1.6973150968552, .60597640275955, .60585051774979, .62808901071548, .23242343962193, 4.7459564208984, 4.7459564208984, 3.0897438526154, 3.0483965873718, 4.2380628585815, 3.7845225334167, .02652905881405, .02652905881405, .01329397037625, .01329396758229, .01330873556435, .00667654490098, .92601269483566, .92601269483566, .60273587703705, .60233747959137, .64300429821014, .44648909568787, .8528453707695, .8528453707695, .72065913677216, .71955502033234, .7962681055069, .7434441447258, .22364875674248, .22364875674248, .16446639597416, .16445553302765, .16780391335487, .12590345740318, -.26983660459518, -.26983660459518, -.1828535348177, -.18284019827843, -.1791032999754, -.11887931078672, -.63495022058487, -.63495022058487, -.53598040342331, -.53542107343674, -.49657794833183, -.38836058974266, -1.7504912614822, -1.7504912614822, -1.2204585075378, -1.2154930830002, -1.0554916858673, -.63642859458923, -1.862185716629, -1.862185716629, -1.2788465023041, -1.2732635736465, -1.1007128953934, -.65061664581299, -2.5617923736572, -2.5617923736572, -1.617108464241, -1.6071890592575, -1.3574055433273, -.71924245357513, -1.6135795116425, -1.6135795116425, -1.1469231843948, -1.1426799297333, -.99809640645981, -.61738300323486, .22420094907284, .22420094907284, .24363535642624, .24356025457382, .25454398989677, .28899359703064, .06624934077263, .06624934077263, .06777309626341, .06777160614729, .06855925172567, .07094971090555, .43318045139313, .43318045139313, .51954871416092, .51871728897095, .57536894083023, .76422989368439, -.89140218496323, -.89140218496323, -.7140833735466, -.7128586769104, -.64815932512283, -.47129172086716]).reshape(17, 6) resids_colnames = ['score_factor', 'resid_response', 'resid_anscombe', 'resid_deviance', 'resid_pearson', 'resid_working'] resids_rownames = ['r'+str(n) for n in range(1, 18)] results_poisson_fweight_nonrobust = ParamsTableTestBunch( params_table=params_table, params_table_colnames=params_table_colnames, params_table_rownames=params_table_rownames, cov=cov, cov_colnames=cov_colnames, cov_rownames=cov_rownames, infocrit=infocrit, infocrit_colnames=infocrit_colnames, infocrit_rownames=infocrit_rownames, predicted=predicted, predicted_colnames=predicted_colnames, predicted_rownames=predicted_rownames, resids=resids, resids_colnames=resids_colnames, resids_rownames=resids_rownames, **est ) est = dict( deviance=12.02863083186947, dispers=1.202863083186947, deviance_s=12.02863083186947, dispers_s=1.202863083186947, deviance_p=15.48630027479802, dispers_p=1.548630027479802, deviance_ps=15.48630027479802, dispers_ps=1.548630027479802, bic=-16.30350260869269, nbml=0, N=17, ic=3, k=7, k_eq=1, k_dv=1, converged=1, k_autoCns=0, ll=-27.28727618329841, chi2=94.62492461274286, p=3.30927661191e-18, rc=0, aic=4.033797198035106, rank=7, canonical=1, power=0, df_m=6, df=10, vf=1, phi=1, k_eq_model=0, properties="b V", depvar="executions", which="max", technique="nr", singularHmethod="m-marquardt", ml_method="e2", crittype="log likelihood", user="glim_lf", title="Generalized linear models", opt="moptimize", chi2type="Wald", wtype="aweight", wexp="= fweight", link="glim_l03", varfunc="glim_v3", m="1", a="1", oim="oim", opt1="ML", varfuncf="u", varfunct="Poisson", linkf="ln(u)", linkt="Log", vce="oim", vcetype="OIM", hac_lag="15", marginsok="default", marginsnotok="stdp Anscombe Cooksd Deviance Hat Likelihood Pearson Response Score Working ADJusted STAndardized STUdentized MODified", # noqa:E501 predict="glim_p", cmd="glm", cmdline="glm executions income perpoverty perblack LN_VC100k96 south degree [aweight=fweight], family(poisson)", # noqa:E501 ) params_table = np.array([ .00025343868829, .00005594520811, 4.5301232557793, 5.894928560e-06, .00014378809529, .00036308928129, np.nan, 1.9599639845401, 0, .09081422305585, .09018031800722, 1.0070293059798, .31392069129295, -.08593595235267, .26756439846436, np.nan, 1.9599639845401, 0, -.09416451429381, .02501975991718, -3.7636058301716, .00016748080115, -.14320234263332, -.04512668595429, np.nan, 1.9599639845401, 0, .27652273809507, .53816281293549, .51382728692594, .60737274844619, -.77825699307725, 1.3313024692674, np.nan, 1.9599639845401, 0, 2.239890838384, .50630271729905, 4.424015044464, 9.688326910e-06, 1.2475557472031, 3.2322259295649, np.nan, 1.9599639845401, 0, -18.842583191417, 5.2065333302747, -3.6190267105084, .00029571311817, -29.047201003062, -8.6379653797707, np.nan, 1.9599639845401, 0, -6.5630017977417, 4.5075460479893, -1.4560032727052, .14539171490364, -15.397629710457, 2.2716261149733, np.nan, 1.9599639845401, 0]).reshape(7, 9) params_table_colnames = 'b se z pvalue ll ul df crit eform'.split() params_table_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov = np.array([ 3.129866310e-09, 2.467206465e-06, 1.706246053e-07, -3.176637541e-07, -6.238332985e-06, -.00010954424563, -.000120350676, 2.467206465e-06, .00813248975588, .00032161167774, -.00128998199687, -.01831266258952, .14012008775466, -.21862639048575, 1.706246053e-07, .00032161167774, .00062598838631, -.00690661599067, -.00759732903266, -.03651168891971, .03689627396044, -3.176637541e-07, -.00128998199687, -.00690661599067, .28961921322663, .04817967329131, .52320524326798, -1.8618850102603, -6.238332985e-06, -.01831266258952, -.00759732903266, .04817967329131, .2563424415444, .86118714295143, -.05787604759173, -.00010954424563, .14012008775466, -.03651168891971, .52320524326798, .86118714295143, 27.107989319261, -7.0873136260377, -.000120350676, -.21862639048575, .03689627396044, -1.8618850102603, -.05787604759173, -7.0873136260377, 20.317971374744]).reshape(7, 7) cov_colnames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] infocrit = np.array([ 17, np.nan, -27.287276183298, 7, 68.574552366597, 74.40704577499]) infocrit_colnames = 'N ll0 ll df AIC BIC'.split() infocrit_rownames = '.'.split() predicted = np.array([ 34.815238952637, .23209382593632, 1.8754115104675, 7.3026847839355, .45640400052071, 1.521183013916, 1.2540435791016, .36330956220627, .16552601754665, 3.9734709262848, .34751656651497, .47986721992493, 2.0739872455597, .34389564394951, .2452784627676, 1.1471545696259, .63293009996414, .45955070853233, 1.7763512134552, .38466224074364, .2628378868103, 2.2698366641998, .35724925994873, .28969252109528, 1.6349502801895, .38501682877541, .24236169457436, 2.7504913806915, .55104273557663, .83518141508102, 2.862185716629, .55352979898453, .87696009874344, 3.5617923736572, .54546248912811, 1.0597376823425, 2.6135795116425, .41179683804512, .44320201873779, .775799036026, .5664399266243, .24891836941242, .93375068902969, .40948873758316, .15657225251198, .56681954860687, .40214782953262, .09166768193245, 1.8914022445679, .30498126149178, .17592607438564]).reshape(17, 3) predicted_colnames = 'predict_mu predict_linpred_std predict_hat'.split() predicted_rownames = ['r'+str(n) for n in range(1, 18)] resids = np.array([ 2.1847612857819, 2.1847612857819, .36650228500366, .36649596691132, .3702706694603, .06275302171707, 1.6973150968552, 1.6973150968552, .60597640275955, .60585051774979, .62808901071548, .23242343962193, 4.7459564208984, 4.7459564208984, 3.0897438526154, 3.0483965873718, 4.2380628585815, 3.7845225334167, .02652905881405, .02652905881405, .01329397037625, .01329396758229, .01330873556435, .00667654490098, .92601269483566, .92601269483566, .60273587703705, .60233747959137, .64300429821014, .44648909568787, .8528453707695, .8528453707695, .72065913677216, .71955502033234, .7962681055069, .7434441447258, .22364875674248, .22364875674248, .16446639597416, .16445553302765, .16780391335487, .12590345740318, -.26983660459518, -.26983660459518, -.1828535348177, -.18284019827843, -.1791032999754, -.11887931078672, -.63495022058487, -.63495022058487, -.53598040342331, -.53542107343674, -.49657794833183, -.38836058974266, -1.7504912614822, -1.7504912614822, -1.2204585075378, -1.2154930830002, -1.0554916858673, -.63642859458923, -1.862185716629, -1.862185716629, -1.2788465023041, -1.2732635736465, -1.1007128953934, -.65061664581299, -2.5617923736572, -2.5617923736572, -1.617108464241, -1.6071890592575, -1.3574055433273, -.71924245357513, -1.6135795116425, -1.6135795116425, -1.1469231843948, -1.1426799297333, -.99809640645981, -.61738300323486, .22420094907284, .22420094907284, .24363535642624, .24356025457382, .25454398989677, .28899359703064, .06624934077263, .06624934077263, .06777309626341, .06777160614729, .06855925172567, .07094971090555, .43318045139313, .43318045139313, .51954871416092, .51871728897095, .57536894083023, .76422989368439, -.89140218496323, -.89140218496323, -.7140833735466, -.7128586769104, -.64815932512283, -.47129172086716]).reshape(17, 6) resids_colnames = ['score_factor', 'resid_response', 'resid_anscombe', 'resid_deviance', 'resid_pearson', 'resid_working'] resids_rownames = ['r'+str(n) for n in range(1, 18)] results_poisson_aweight_nonrobust = ParamsTableTestBunch( params_table=params_table, params_table_colnames=params_table_colnames, params_table_rownames=params_table_rownames, cov=cov, cov_colnames=cov_colnames, cov_rownames=cov_rownames, infocrit=infocrit, infocrit_colnames=infocrit_colnames, infocrit_rownames=infocrit_rownames, predicted=predicted, predicted_colnames=predicted_colnames, predicted_rownames=predicted_rownames, resids=resids, resids_colnames=resids_colnames, resids_rownames=resids_rownames, **est ) est = dict( deviance=23.34969514421719, dispers=2.33496951442172, deviance_s=23.34969514421719, dispers_s=2.33496951442172, deviance_p=30.06164170990202, dispers_p=3.006164170990202, deviance_ps=30.06164170990202, dispers_ps=3.006164170990202, bic=-4.982438296344967, nbml=0, N=17, ic=3, k=7, k_eq=1, k_dv=1, converged=1, k_autoCns=0, ll=-52.96941847346162, chi2=356.6637749656061, p=5.72458312679e-74, rc=0, aic=7.055225702760191, rank=7, canonical=1, power=0, df_m=6, df=10, vf=1, phi=1, k_eq_model=0, properties="b V", depvar="executions", which="max", technique="nr", singularHmethod="m-marquardt", ml_method="e2", crittype="log pseudolikelihood", user="glim_lf", title="Generalized linear models", opt="moptimize", chi2type="Wald", wtype="pweight", wexp="= fweight", link="glim_l03", varfunc="glim_v3", m="1", a="1", oim="oim", opt1="ML", varfuncf="u", varfunct="Poisson", linkf="ln(u)", linkt="Log", vcetype="Robust", hac_lag="15", marginsok="default", marginsnotok="stdp Anscombe Cooksd Deviance Hat Likelihood Pearson Response Score Working ADJusted STAndardized STUdentized MODified", # noqa:E501 predict="glim_p", cmd="glm", cmdline="glm executions income perpoverty perblack LN_VC100k96 south degree [pweight=fweight], family(poisson)", # noqa:E501 ) params_table = np.array([ .00025343868829, .0000298866597, 8.4799937786829, 2.252059827e-17, .00019486191167, .00031201546491, np.nan, 1.9599639845401, 0, .09081422305585, .08414617969117, 1.0792435662456, .28047916301946, -.07410925857549, .25573770468718, np.nan, 1.9599639845401, 0, -.09416451429381, .01946961498728, -4.8364856909253, 1.321547815e-06, -.13232425846174, -.05600477012587, np.nan, 1.9599639845401, 0, .27652273809506, .36112179485191, .76573261995571, .44383541350407, -.43126297384714, .98430845003726, np.nan, 1.9599639845401, 0, 2.239890838384, .43098853454849, 5.1971007551989, 2.024206636e-07, 1.3951688329193, 3.0846128438487, np.nan, 1.9599639845401, 0, -18.842583191417, 4.5147658917489, -4.1735460139479, .00002998950578, -27.691361737874, -9.9938046449589, np.nan, 1.9599639845401, 0, -6.5630017977416, 3.3999612612355, -1.930316639948, .0535676165153, -13.226803418595, .10079982311137, np.nan, 1.9599639845401, 0]).reshape(7, 9) params_table_colnames = 'b se z pvalue ll ul df crit eform'.split() params_table_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov = np.array([ 8.932124278e-10, 1.512127962e-06, 1.877263788e-07, -4.562869239e-06, -2.023379829e-06, -.00001228516761, -.00002423071544, 1.512127962e-06, .00708057955662, .00028427703202, -.0019549511748, -.00596332288528, .20022061835302, -.18678265108673, 1.877263788e-07, .00028427703202, .00037906590775, -.00453407701816, -.00623061980467, -.04659404972535, .02694184589715, -4.562869239e-06, -.0019549511748, -.00453407701816, .13040895071706, .0836259691825, .89260578257395, -.82275604425197, -2.023379829e-06, -.00596332288528, -.00623061980467, .0836259691825, .18575111691225, 1.0698498854979, -.64859219982217, -.00001228516761, .20022061835302, -.04659404972535, .89260578257395, 1.0698498854979, 20.383111057299, -12.482192460755, -.00002423071544, -.18678265108673, .02694184589715, -.82275604425197, -.64859219982217, -12.482192460755, 11.559736577902]).reshape(7, 7) cov_colnames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] infocrit = np.array([ 17, np.nan, -52.969418473462, 7, 119.93883694692, 125.77133035532]) infocrit_colnames = 'N ll0 ll df AIC BIC'.split() infocrit_rownames = '.'.split() predicted = np.array([ 34.815238952637, .06858423352242, 7.3026847839355, .25687274336815, 1.2540435791016, .41320022940636, 3.9734709262848, .16020278632641, 2.0739872455597, .22170753777027, 1.1471545696259, .51121062040329, 1.7763512134552, .2167394310236, 2.2698366641998, .2456086575985, 1.6349502801895, .25546172261238, 2.7504913806915, .4417819082737, 2.862185716629, .61734634637833, 3.5617923736572, .51518148183823, 2.6135795116425, .34006628394127, .775799036026, .292076587677, .93375068902969, .39795544743538, .56681954860687, .31529840826988, 1.8914022445679, .26116076111794]).reshape(17, 2) predicted_colnames = 'predict_mu predict_linpred_std'.split() predicted_rownames = ['r'+str(n) for n in range(1, 18)] resids = np.array([ 2.1847612857819, 2.1847612857819, .36650228500366, .36649596691132, .3702706694603, .06275302171707, 1.6973150968552, 1.6973150968552, .60597640275955, .60585051774979, .62808901071548, .23242343962193, 4.7459564208984, 4.7459564208984, 3.0897438526154, 3.0483965873718, 4.2380628585815, 3.7845225334167, .02652905881405, .02652905881405, .01329397037625, .01329396758229, .01330873556435, .00667654490098, .92601269483566, .92601269483566, .60273587703705, .60233747959137, .64300429821014, .44648909568787, .8528453707695, .8528453707695, .72065913677216, .71955502033234, .7962681055069, .7434441447258, .22364875674248, .22364875674248, .16446639597416, .16445553302765, .16780391335487, .12590345740318, -.26983660459518, -.26983660459518, -.1828535348177, -.18284019827843, -.1791032999754, -.11887931078672, -.63495022058487, -.63495022058487, -.53598040342331, -.53542107343674, -.49657794833183, -.38836058974266, -1.7504912614822, -1.7504912614822, -1.2204585075378, -1.2154930830002, -1.0554916858673, -.63642859458923, -1.862185716629, -1.862185716629, -1.2788465023041, -1.2732635736465, -1.1007128953934, -.65061664581299, -2.5617923736572, -2.5617923736572, -1.617108464241, -1.6071890592575, -1.3574055433273, -.71924245357513, -1.6135795116425, -1.6135795116425, -1.1469231843948, -1.1426799297333, -.99809640645981, -.61738300323486, .22420094907284, .22420094907284, .24363535642624, .24356025457382, .25454398989677, .28899359703064, .06624934077263, .06624934077263, .06777309626341, .06777160614729, .06855925172567, .07094971090555, .43318045139313, .43318045139313, .51954871416092, .51871728897095, .57536894083023, .76422989368439, -.89140218496323, -.89140218496323, -.7140833735466, -.7128586769104, -.64815932512283, -.47129172086716]).reshape(17, 6) resids_colnames = ['score_factor', 'resid_response', 'resid_anscombe', 'resid_deviance', 'resid_pearson', 'resid_working'] resids_rownames = ['r'+str(n) for n in range(1, 18)] results_poisson_pweight_nonrobust = ParamsTableTestBunch( params_table=params_table, params_table_colnames=params_table_colnames, params_table_rownames=params_table_rownames, cov=cov, cov_colnames=cov_colnames, cov_rownames=cov_rownames, infocrit=infocrit, infocrit_colnames=infocrit_colnames, infocrit_rownames=infocrit_rownames, predicted=predicted, predicted_colnames=predicted_colnames, predicted_rownames=predicted_rownames, resids=resids, resids_colnames=resids_colnames, resids_rownames=resids_rownames, **est ) est = dict( k_eq_model=0, phi=1, vf=1, df=10, df_m=6, power=0, canonical=1, rank=7, aic=4.579685683305704, rc=0, p=5.09268495340e-76, chi2=366.2131475852884, ll=-31.92732830809848, k_autoCns=0, converged=1, k_dv=1, k_eq=1, k=7, ic=3, N=17, nbml=0, bic=-9.740492454486454, dispers_ps=2.475374834715614, deviance_ps=24.75374834715614, dispers_p=2.475374834715614, deviance_p=24.75374834715614, dispers_s=1.859164098607571, deviance_s=18.59164098607571, dispers=1.859164098607571, deviance=18.59164098607571, cmdline="glm executions income perpoverty perblack LN_VC100k96 south degree, family(poisson) vce(robust)", # noqa:E501 cmd="glm", predict="glim_p", marginsnotok="stdp Anscombe Cooksd Deviance Hat Likelihood Pearson Response Score Working ADJusted STAndardized STUdentized MODified", # noqa:E501 marginsok="default", hac_lag="15", vcetype="Robust", vce="robust", linkt="Log", linkf="ln(u)", varfunct="Poisson", varfuncf="u", opt1="ML", oim="oim", a="1", m="1", varfunc="glim_v3", link="glim_l03", chi2type="Wald", opt="moptimize", title="Generalized linear models", user="glim_lf", crittype="log pseudolikelihood", ml_method="e2", singularHmethod="m-marquardt", technique="nr", which="max", depvar="executions", properties="b V", ) params_table = np.array([ .00026110166569, .00003534474167, 7.3872845963787, 1.498576223e-13, .00019182724497, .0003303760864, np.nan, 1.9599639845401, 0, .07781804809828, .09819599835909, .79247677500784, .42808272865983, -.11464257211148, .27027866830805, np.nan, 1.9599639845401, 0, -.09493110013466, .01944446025221, -4.8821668950083, 1.049263903e-06, -.13304154192782, -.0568206583415, np.nan, 1.9599639845401, 0, .29693462055586, .34917491559373, .85038932436186, .39510866948496, -.38743563831266, .98130487942439, np.nan, 1.9599639845401, 0, 2.3011832004524, .45717041903387, 5.0335347709405, 4.815174289e-07, 1.405145644349, 3.1972207565559, np.nan, 1.9599639845401, 0, -18.722067603077, 4.5006120067298, -4.1598937155841, .00003183957242, -27.543105044656, -9.9010301614985, np.nan, 1.9599639845401, 0, -6.8014789919532, 3.48445447794, -1.9519494471841, .05094420680386, -13.630884274485, .02792629057847, np.nan, 1.9599639845401, 0]).reshape(7, 9) params_table_colnames = 'b se z pvalue ll ul df crit eform'.split() params_table_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov = np.array([ 1.249250764e-09, 2.158351725e-06, 1.068227835e-07, -5.170410321e-06, -5.047866044e-07, -.00001662944527, -.00004339679838, 2.158351725e-06, .00964245409374, .00008635335196, -.00640596402935, -.00524426268669, .23390140895418, -.22653903184676, 1.068227835e-07, .00008635335196, .0003780870345, -.00382751790532, -.0064534643179, -.05137117620883, .02948709519544, -5.170410321e-06, -.00640596402935, -.00382751790532, .12192312167989, .0907733380116, .89729289134262, -.69004336039169, -5.047866044e-07, -.00524426268669, -.0064534643179, .0907733380116, .20900479203961, .93952111535021, -.75843860743141, -.00001662944527, .23390140895418, -.05137117620883, .89729289134262, .93952111535021, 20.25550843512, -12.691830440798, -.00004339679838, -.22653903184676, .02948709519544, -.69004336039169, -.75843860743141, -12.691830440798, 12.141423008836]).reshape(7, 7) cov_colnames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] infocrit = np.array([ 17, np.nan, -31.927328308098, 7, 77.854656616197, 83.68715002459]) infocrit_colnames = 'N ll0 ll df AIC BIC'.split() infocrit_rownames = '.'.split() predicted = np.array([ 35.226364135742, .05631958693266, 8.1965742111206, .14089094102383, 1.3118965625763, .51714926958084, 3.6862981319427, .20286601781845, 2.0823004245758, .27275583148003, 1.0650315284729, .58616667985916, 1.9260421991348, .30098018050194, 2.4171404838562, .34251752495766, 1.8473218679428, .29685723781586, 2.8643238544464, .47364214062691, 3.1211984157562, .72507524490356, 3.338207244873, .54493451118469, 2.5269968509674, .34425318241119, .89725440740585, .37162157893181, .97933322191238, .50227928161621, .53462094068527, .40906101465225, 1.9790935516357, .33805811405182]).reshape(17, 2) predicted_colnames = 'predict_mu predict_linpred_std'.split() predicted_rownames = ['r'+str(n) for n in range(1, 18)] results_poisson_none_hc1 = ParamsTableTestBunch( params_table=params_table, params_table_colnames=params_table_colnames, params_table_rownames=params_table_rownames, cov=cov, cov_colnames=cov_colnames, cov_rownames=cov_rownames, infocrit=infocrit, infocrit_colnames=infocrit_colnames, infocrit_rownames=infocrit_rownames, predicted=predicted, predicted_colnames=predicted_colnames, predicted_rownames=predicted_rownames, **est ) est = dict( k_eq_model=0, phi=1, vf=1, df=26, df_m=6, power=0, canonical=1, rank=7, aic=3.634510210512826, rc=0, p=1.5690245831e-115, chi2=549.7874580263729, ll=-52.96941847346162, k_autoCns=0, converged=1, k_dv=1, k_eq=1, k=7, ic=3, N=33, nbml=0, bic=-67.5595014539113, dispers_ps=1.156216988842385, deviance_ps=30.06164170990202, dispers_p=1.156216988842385, deviance_p=30.06164170990202, dispers_s=.8980651978545075, deviance_s=23.34969514421719, dispers=.8980651978545075, deviance=23.34969514421719, cmdline="glm executions income perpoverty perblack LN_VC100k96 south degree [fweight=fweight], family(poisson) vce(robust)", # noqa:E501 cmd="glm", predict="glim_p", marginsnotok="stdp Anscombe Cooksd Deviance Hat Likelihood Pearson Response Score Working ADJusted STAndardized STUdentized MODified", # noqa:E501 marginsok="default", hac_lag="15", vcetype="Robust", vce="robust", linkt="Log", linkf="ln(u)", varfunct="Poisson", varfuncf="u", opt1="ML", oim="oim", a="1", m="1", varfunc="glim_v3", link="glim_l03", wexp="= fweight", wtype="fweight", chi2type="Wald", opt="moptimize", title="Generalized linear models", user="glim_lf", crittype="log pseudolikelihood", ml_method="e2", singularHmethod="m-marquardt", technique="nr", which="max", depvar="executions", properties="b V", ) params_table = np.array([ .00025343868829, .0000263369674, 9.6229259983619, 6.398464168e-22, .00020181918073, .00030505819585, np.nan, 1.9599639845401, 0, .09081422305585, .07431850776812, 1.2219597215163, .22172285914198, -.05484737555444, .23647582166613, np.nan, 1.9599639845401, 0, -.09416451429381, .01609416304158, -5.8508487860178, 4.890707145e-09, -.12570849421662, -.06262053437099, np.nan, 1.9599639845401, 0, .27652273809506, .34481886883624, .80193621372381, .42258985672342, -.3993098260138, .95235530220392, np.nan, 1.9599639845401, 0, 2.239890838384, .39682271484988, 5.6445630619491, 1.656012749e-08, 1.4621326090308, 3.0176490677372, np.nan, 1.9599639845401, 0, -18.842583191417, 4.1473740870735, -4.5432562377589, 5.539185130e-06, -26.971287032495, -10.713879350338, np.nan, 1.9599639845401, 0, -6.5630017977416, 3.0810023455152, -2.1301515097173, .03315910688542, -12.601655431235, -.52434816424841, np.nan, 1.9599639845401, 0]).reshape(7, 9) params_table_colnames = 'b se z pvalue ll ul df crit eform'.split() params_table_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov = np.array([ 6.936358517e-10, 1.301395377e-06, 1.497821854e-07, -4.758016826e-06, -1.852598001e-06, -6.904571080e-06, -.00001327109619, 1.301395377e-06, .00552324059688, .00014714335792, -.00376147485446, -.00118957690573, .15979100738539, -.13853266210904, 1.497821854e-07, .00014714335792, .00025902208401, -.00418693954572, -.00513741847691, -.03987504442994, .02761179707845, -4.758016826e-06, -.00376147485446, -.00418693954572, .1189000523055, .08682729933237, .80541854027627, -.70545315416752, -1.852598001e-06, -.00118957690573, -.00513741847691, .08682729933237, .15746826702083, 1.1366624064282, -.75098089879076, -6.904571080e-06, .15979100738539, -.03987504442994, .80541854027627, 1.1366624064282, 17.200711818129, -11.062121016981, -.00001327109619, -.13853266210904, .02761179707845, -.70545315416752, -.75098089879076, -11.062121016981, 9.49257545307]).reshape(7, 7) cov_colnames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] infocrit = np.array([ 33, np.nan, -52.969418473462, 7, 119.93883694692, 130.41438987719]) infocrit_colnames = 'N ll0 ll df AIC BIC'.split() infocrit_rownames = '.'.split() predicted = np.array([ 34.815238952637, .06608480215073, 7.3026847839355, .23366995155811, 1.2540435791016, .39606991410255, 3.9734709262848, .12350843846798, 2.0739872455597, .18263976275921, 1.1471545696259, .39735752344131, 1.7763512134552, .17952646315098, 2.2698366641998, .21028706431389, 1.6349502801895, .17675416171551, 2.7504913806915, .42150634527206, 2.862185716629, .58209121227264, 3.5617923736572, .49835306406021, 2.6135795116425, .2456089258194, .775799036026, .23251366615295, .93375068902969, .35320028662682, .56681954860687, .26245352625847, 1.8914022445679, .20374123752117]).reshape(17, 2) predicted_colnames = 'predict_mu predict_linpred_std'.split() predicted_rownames = ['r'+str(n) for n in range(1, 18)] results_poisson_fweight_hc1 = ParamsTableTestBunch( params_table=params_table, params_table_colnames=params_table_colnames, params_table_rownames=params_table_rownames, cov=cov, cov_colnames=cov_colnames, cov_rownames=cov_rownames, infocrit=infocrit, infocrit_colnames=infocrit_colnames, infocrit_rownames=infocrit_rownames, predicted=predicted, predicted_colnames=predicted_colnames, predicted_rownames=predicted_rownames, **est ) est = dict( k_eq_model=0, phi=1, vf=1, df=10, df_m=6, power=0, canonical=1, rank=7, aic=4.033797198035106, rc=0, p=5.72458312675e-74, chi2=356.663774965618, ll=-27.28727618329841, k_autoCns=0, converged=1, k_dv=1, k_eq=1, k=7, ic=3, N=17, nbml=0, bic=-16.30350260869269, dispers_ps=1.548630027479802, deviance_ps=15.48630027479802, dispers_p=1.548630027479802, deviance_p=15.48630027479802, dispers_s=1.202863083186947, deviance_s=12.02863083186947, dispers=1.202863083186947, deviance=12.02863083186947, cmdline="glm executions income perpoverty perblack LN_VC100k96 south degree [aweight=fweight], family(poisson) vce(robust)", # noqa:E501 cmd="glm", predict="glim_p", marginsnotok="stdp Anscombe Cooksd Deviance Hat Likelihood Pearson Response Score Working ADJusted STAndardized STUdentized MODified", # noqa:E501 marginsok="default", hac_lag="15", vcetype="Robust", vce="robust", linkt="Log", linkf="ln(u)", varfunct="Poisson", varfuncf="u", opt1="ML", oim="oim", a="1", m="1", varfunc="glim_v3", link="glim_l03", wexp="= fweight", wtype="aweight", chi2type="Wald", opt="moptimize", title="Generalized linear models", user="glim_lf", crittype="log pseudolikelihood", ml_method="e2", singularHmethod="m-marquardt", technique="nr", which="max", depvar="executions", properties="b V", ) params_table = np.array([ .00025343868829, .0000298866597, 8.4799937786833, 2.252059827e-17, .00019486191167, .00031201546491, np.nan, 1.9599639845401, 0, .09081422305585, .08414617969118, 1.0792435662455, .28047916301948, -.0741092585755, .25573770468719, np.nan, 1.9599639845401, 0, -.09416451429381, .01946961498728, -4.8364856909248, 1.321547815e-06, -.13232425846174, -.05600477012587, np.nan, 1.9599639845401, 0, .27652273809507, .36112179485206, .76573261995541, .44383541350425, -.43126297384744, .98430845003758, np.nan, 1.9599639845401, 0, 2.239890838384, .4309885345485, 5.1971007551988, 2.024206636e-07, 1.3951688329193, 3.0846128438488, np.nan, 1.9599639845401, 0, -18.842583191417, 4.5147658917496, -4.1735460139472, .00002998950578, -27.691361737876, -9.9938046449574, np.nan, 1.9599639845401, 0, -6.5630017977417, 3.3999612612367, -1.9303166399474, .05356761651539, -13.226803418597, .10079982311369, np.nan, 1.9599639845401, 0]).reshape(7, 9) params_table_colnames = 'b se z pvalue ll ul df crit eform'.split() params_table_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov = np.array([ 8.932124278e-10, 1.512127962e-06, 1.877263788e-07, -4.562869239e-06, -2.023379829e-06, -.00001228516761, -.00002423071544, 1.512127962e-06, .00708057955662, .00028427703202, -.00195495117479, -.00596332288528, .2002206183531, -.1867826510868, 1.877263788e-07, .00028427703202, .00037906590775, -.00453407701816, -.00623061980468, -.04659404972537, .02694184589718, -4.562869239e-06, -.00195495117479, -.00453407701816, .13040895071718, .08362596918255, .89260578257483, -.82275604425296, -2.023379829e-06, -.00596332288528, -.00623061980468, .08362596918255, .18575111691226, 1.0698498854982, -.64859219982256, -.00001228516761, .2002206183531, -.04659404972537, .89260578257483, 1.0698498854982, 20.383111057306, -12.482192460764, -.00002423071544, -.1867826510868, .02694184589718, -.82275604425296, -.64859219982256, -12.482192460764, 11.55973657791]).reshape(7, 7) cov_colnames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] infocrit = np.array([ 17, np.nan, -27.287276183298, 7, 68.574552366597, 74.40704577499]) infocrit_colnames = 'N ll0 ll df AIC BIC'.split() infocrit_rownames = '.'.split() predicted = np.array([ 34.815238952637, .06858423352242, 7.3026847839355, .25687274336815, 1.2540435791016, .41320022940636, 3.9734709262848, .16020278632641, 2.0739872455597, .22170753777027, 1.1471545696259, .51121062040329, 1.7763512134552, .2167394310236, 2.2698366641998, .2456086575985, 1.6349502801895, .25546172261238, 2.7504913806915, .4417819082737, 2.862185716629, .61734634637833, 3.5617923736572, .51518148183823, 2.6135795116425, .34006628394127, .775799036026, .292076587677, .93375068902969, .39795544743538, .56681954860687, .31529840826988, 1.8914022445679, .26116076111794]).reshape(17, 2) predicted_colnames = 'predict_mu predict_linpred_std'.split() predicted_rownames = ['r'+str(n) for n in range(1, 18)] results_poisson_aweight_hc1 = ParamsTableTestBunch( params_table=params_table, params_table_colnames=params_table_colnames, params_table_rownames=params_table_rownames, cov=cov, cov_colnames=cov_colnames, cov_rownames=cov_rownames, infocrit=infocrit, infocrit_colnames=infocrit_colnames, infocrit_rownames=infocrit_rownames, predicted=predicted, predicted_colnames=predicted_colnames, predicted_rownames=predicted_rownames, **est ) est = dict( k_eq_model=0, phi=1, vf=1, df=10, df_m=6, power=0, canonical=1, rank=7, aic=7.055225702760191, rc=0, p=5.72458312679e-74, chi2=356.6637749656061, ll=-52.96941847346162, k_autoCns=0, converged=1, k_dv=1, k_eq=1, k=7, ic=3, N=17, nbml=0, bic=-4.982438296344967, dispers_ps=3.006164170990202, deviance_ps=30.06164170990202, dispers_p=3.006164170990202, deviance_p=30.06164170990202, dispers_s=2.33496951442172, deviance_s=23.34969514421719, dispers=2.33496951442172, deviance=23.34969514421719, cmdline="glm executions income perpoverty perblack LN_VC100k96 south degree [pweight=fweight], family(poisson) vce(robust)", # noqa:E501 cmd="glm", predict="glim_p", marginsnotok="stdp Anscombe Cooksd Deviance Hat Likelihood Pearson Response Score Working ADJusted STAndardized STUdentized MODified", # noqa:E501 marginsok="default", hac_lag="15", vcetype="Robust", vce="robust", linkt="Log", linkf="ln(u)", varfunct="Poisson", varfuncf="u", opt1="ML", oim="oim", a="1", m="1", varfunc="glim_v3", link="glim_l03", wexp="= fweight", wtype="pweight", chi2type="Wald", opt="moptimize", title="Generalized linear models", user="glim_lf", crittype="log pseudolikelihood", ml_method="e2", singularHmethod="m-marquardt", technique="nr", which="max", depvar="executions", properties="b V", ) params_table = np.array([ .00025343868829, .0000298866597, 8.4799937786829, 2.252059827e-17, .00019486191167, .00031201546491, np.nan, 1.9599639845401, 0, .09081422305585, .08414617969117, 1.0792435662456, .28047916301946, -.07410925857549, .25573770468718, np.nan, 1.9599639845401, 0, -.09416451429381, .01946961498728, -4.8364856909253, 1.321547815e-06, -.13232425846174, -.05600477012587, np.nan, 1.9599639845401, 0, .27652273809506, .36112179485191, .76573261995571, .44383541350407, -.43126297384714, .98430845003726, np.nan, 1.9599639845401, 0, 2.239890838384, .43098853454849, 5.1971007551989, 2.024206636e-07, 1.3951688329193, 3.0846128438487, np.nan, 1.9599639845401, 0, -18.842583191417, 4.5147658917489, -4.1735460139479, .00002998950578, -27.691361737874, -9.9938046449589, np.nan, 1.9599639845401, 0, -6.5630017977416, 3.3999612612355, -1.930316639948, .0535676165153, -13.226803418595, .10079982311137, np.nan, 1.9599639845401, 0]).reshape(7, 9) params_table_colnames = 'b se z pvalue ll ul df crit eform'.split() params_table_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov = np.array([ 8.932124278e-10, 1.512127962e-06, 1.877263788e-07, -4.562869239e-06, -2.023379829e-06, -.00001228516761, -.00002423071544, 1.512127962e-06, .00708057955662, .00028427703202, -.0019549511748, -.00596332288528, .20022061835302, -.18678265108673, 1.877263788e-07, .00028427703202, .00037906590775, -.00453407701816, -.00623061980467, -.04659404972535, .02694184589715, -4.562869239e-06, -.0019549511748, -.00453407701816, .13040895071706, .0836259691825, .89260578257395, -.82275604425197, -2.023379829e-06, -.00596332288528, -.00623061980467, .0836259691825, .18575111691225, 1.0698498854979, -.64859219982217, -.00001228516761, .20022061835302, -.04659404972535, .89260578257395, 1.0698498854979, 20.383111057299, -12.482192460755, -.00002423071544, -.18678265108673, .02694184589715, -.82275604425197, -.64859219982217, -12.482192460755, 11.559736577902]).reshape(7, 7) cov_colnames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] infocrit = np.array([ 17, np.nan, -52.969418473462, 7, 119.93883694692, 125.77133035532]) infocrit_colnames = 'N ll0 ll df AIC BIC'.split() infocrit_rownames = '.'.split() predicted = np.array([ 34.815238952637, .06858423352242, 7.3026847839355, .25687274336815, 1.2540435791016, .41320022940636, 3.9734709262848, .16020278632641, 2.0739872455597, .22170753777027, 1.1471545696259, .51121062040329, 1.7763512134552, .2167394310236, 2.2698366641998, .2456086575985, 1.6349502801895, .25546172261238, 2.7504913806915, .4417819082737, 2.862185716629, .61734634637833, 3.5617923736572, .51518148183823, 2.6135795116425, .34006628394127, .775799036026, .292076587677, .93375068902969, .39795544743538, .56681954860687, .31529840826988, 1.8914022445679, .26116076111794]).reshape(17, 2) predicted_colnames = 'predict_mu predict_linpred_std'.split() predicted_rownames = ['r'+str(n) for n in range(1, 18)] results_poisson_pweight_hc1 = ParamsTableTestBunch( params_table=params_table, params_table_colnames=params_table_colnames, params_table_rownames=params_table_rownames, cov=cov, cov_colnames=cov_colnames, cov_rownames=cov_rownames, infocrit=infocrit, infocrit_colnames=infocrit_colnames, infocrit_rownames=infocrit_rownames, predicted=predicted, predicted_colnames=predicted_colnames, predicted_rownames=predicted_rownames, **est ) est = dict( k_eq_model=0, vf=1, df=10, df_m=6, power=0, canonical=1, rank=7, aic=4.579685683305704, rc=0, p=4.1950730971e-123, chi2=584.908728768987, ll=-31.92732830809848, N_clust=9, k_autoCns=0, converged=1, k_dv=1, k_eq=1, k=7, ic=3, N=17, nbml=0, bic=-9.740492454486454, dispers_ps=2.475374834715614, deviance_ps=24.75374834715614, dispers_p=2.475374834715614, deviance_p=24.75374834715614, dispers_s=1.859164098607571, deviance_s=18.59164098607571, dispers=1.859164098607571, deviance=18.59164098607571, phi=1, cmdline="glm executions income perpoverty perblack LN_VC100k96 south degree, family(poisson) vce(cluster id)", # noqa:E501 cmd="glm", predict="glim_p", marginsnotok="stdp Anscombe Cooksd Deviance Hat Likelihood Pearson Response Score Working ADJusted STAndardized STUdentized MODified", # noqa:E501 marginsok="default", hac_lag="15", vcetype="Robust", vce="cluster", linkt="Log", linkf="ln(u)", varfunct="Poisson", varfuncf="u", opt1="ML", clustvar="id", oim="oim", a="1", m="1", varfunc="glim_v3", link="glim_l03", chi2type="Wald", opt="moptimize", title="Generalized linear models", user="glim_lf", crittype="log pseudolikelihood", ml_method="e2", singularHmethod="m-marquardt", technique="nr", which="max", depvar="executions", properties="b V", ) params_table = np.array([ .00026110166569, .00004098448535, 6.3707440379489, 1.881133617e-10, .00018077355048, .0003414297809, np.nan, 1.9599639845401, 0, .07781804809828, .11602998752167, .67067186475175, .50242959011024, -.14959654857083, .3052326447674, np.nan, 1.9599639845401, 0, -.09493110013466, .02432927475974, -3.9019288931601, .00009542919351, -.14261560243373, -.04724659783559, np.nan, 1.9599639845401, 0, .29693462055586, .31774950884716, .93449277587615, .35004976070702, -.32584297288986, .91971221400158, np.nan, 1.9599639845401, 0, 2.3011832004524, .54874508731474, 4.1935376801516, .00002746374324, 1.2256625926223, 3.3767038082826, np.nan, 1.9599639845401, 0, -18.722067603077, 2.8106198749749, -6.6611880780372, 2.716227723e-11, -24.230781332261, -13.213353873894, np.nan, 1.9599639845401, 0, -6.8014789919532, 3.1571598785659, -2.1543029981246, .03121641791743, -12.989398647377, -.61355933652912, np.nan, 1.9599639845401, 0]).reshape(7, 9) params_table_colnames = 'b se z pvalue ll ul df crit eform'.split() params_table_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov = np.array([ 1.679728039e-09, 4.034336761e-06, 1.735749447e-07, -5.093610363e-06, -4.552211884e-06, .00001563785418, -.00009230028034, 4.034336761e-06, .01346295800428, .00110922683659, -.01950093608551, -.02957572460439, .08545644123676, -.23518641056668, 1.735749447e-07, .00110922683659, .00059191361033, -.00720622811203, -.01195031391163, -.04317371228367, .03351736744645, -5.093610363e-06, -.01950093608551, -.00720622811203, .10096475037261, .13375578883899, .49763538443989, -.27357574414228, -4.552211884e-06, -.02957572460439, -.01195031391163, .13375578883899, .30112117085206, .65342245458316, -.47102547759356, .00001563785418, .08545644123676, -.04317371228367, .49763538443989, .65342245458316, 7.8995840816039, -6.5824964755966, -.00009230028034, -.23518641056668, .03351736744645, -.27357574414228, -.47102547759356, -6.5824964755966, 9.9676584988266]).reshape(7, 7) cov_colnames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] infocrit = np.array([ 17, np.nan, -31.927328308098, 7, 77.854656616197, 83.68715002459]) infocrit_colnames = 'N ll0 ll df AIC BIC'.split() infocrit_rownames = '.'.split() predicted = np.array([ 35.226364135742, .05941947177052, 8.1965742111206, .09018591046333, 1.3118965625763, .53127920627594, 3.6862981319427, .23996050655842, 2.0823004245758, .33554902672768, 1.0650315284729, .53513532876968, 1.9260421991348, .32360115647316, 2.4171404838562, .33078169822693, 1.8473218679428, .32581362128258, 2.8643238544464, .46489810943604, 3.1211984157562, .71297109127045, 3.338207244873, .58515930175781, 2.5269968509674, .42410242557526, .89725440740585, .40493285655975, .97933322191238, .5560839176178, .53462094068527, .419488966465, 1.9790935516357, .3438538312912]).reshape(17, 2) predicted_colnames = 'predict_mu predict_linpred_std'.split() predicted_rownames = ['r'+str(n) for n in range(1, 18)] results_poisson_none_clu1 = ParamsTableTestBunch( params_table=params_table, params_table_colnames=params_table_colnames, params_table_rownames=params_table_rownames, cov=cov, cov_colnames=cov_colnames, cov_rownames=cov_rownames, infocrit=infocrit, infocrit_colnames=infocrit_colnames, infocrit_rownames=infocrit_rownames, predicted=predicted, predicted_colnames=predicted_colnames, predicted_rownames=predicted_rownames, **est ) est = dict( k_eq_model=0, vf=1, df=26, df_m=6, power=0, canonical=1, rank=7, aic=3.634510210512826, rc=0, p=6.87057569032e-91, chi2=435.380362705941, ll=-52.96941847346162, N_clust=9, k_autoCns=0, converged=1, k_dv=1, k_eq=1, k=7, ic=3, N=33, nbml=0, bic=-67.5595014539113, dispers_ps=1.156216988842385, deviance_ps=30.06164170990202, dispers_p=1.156216988842385, deviance_p=30.06164170990202, dispers_s=.8980651978545075, deviance_s=23.34969514421719, dispers=.8980651978545075, deviance=23.34969514421719, phi=1, cmdline="glm executions income perpoverty perblack LN_VC100k96 south degree [fweight=fweight], family(poisson) vce(cluster id)", # noqa:E501 cmd="glm", predict="glim_p", marginsnotok="stdp Anscombe Cooksd Deviance Hat Likelihood Pearson Response Score Working ADJusted STAndardized STUdentized MODified", # noqa:E501 marginsok="default", hac_lag="15", vcetype="Robust", vce="cluster", linkt="Log", linkf="ln(u)", varfunct="Poisson", varfuncf="u", opt1="ML", clustvar="id", oim="oim", a="1", m="1", varfunc="glim_v3", link="glim_l03", wexp="= fweight", wtype="fweight", chi2type="Wald", opt="moptimize", title="Generalized linear models", user="glim_lf", crittype="log pseudolikelihood", ml_method="e2", singularHmethod="m-marquardt", technique="nr", which="max", depvar="executions", properties="b V", ) params_table = np.array([ .00025343868829, .0000293670276, 8.6300422274613, 6.132932700e-18, .00019588037186, .00031099700472, np.nan, 1.9599639845401, 0, .09081422305585, .09800194027664, .92665739881773, .35410444288802, -.10126605030142, .28289449641311, np.nan, 1.9599639845401, 0, -.09416451429381, .02511206083893, -3.7497724658197, .00017699509401, -.14338324911569, -.04494577947193, np.nan, 1.9599639845401, 0, .27652273809506, .36749499886987, .75245306451906, .45177864537662, -.44375422418847, .99679970037859, np.nan, 1.9599639845401, 0, 2.239890838384, .51564197481271, 4.343887712395, .00001399830855, 1.229251138834, 3.250530537934, np.nan, 1.9599639845401, 0, -18.842583191417, 3.2292740757113, -5.8349284543976, 5.381365332e-09, -25.17184407602, -12.513322306813, np.nan, 1.9599639845401, 0, -6.5630017977416, 3.1938260811459, -2.0549026875586, .03988840483712, -12.822785889672, -.30321770581092, np.nan, 1.9599639845401, 0]).reshape(7, 9) params_table_colnames = 'b se z pvalue ll ul df crit eform'.split() params_table_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov = np.array([ 8.624223101e-10, 2.413510691e-06, 3.123995891e-07, -4.358439015e-06, -8.084672085e-06, -4.785328653e-06, -.00003652286809, 2.413510691e-06, .00960438029799, .00106422375754, -.00911884619892, -.03121758372723, .06803953530989, -.17715756048416, 3.123995891e-07, .00106422375754, .00063061559958, -.00844230553011, -.01177586448603, -.05361546061036, .03844868195577, -4.358439015e-06, -.00911884619892, -.00844230553011, .13505257419436, .14058853110927, .86184257188631, -.74146699290106, -8.084672085e-06, -.03121758372723, -.01177586448603, .14058853110927, .26588664618875, .75712244813913, -.35118919402718, -4.785328653e-06, .06803953530989, -.05361546061036, .86184257188631, .75712244813913, 10.428211056061, -8.3518020608948, -.00003652286809, -.17715756048416, .03844868195577, -.74146699290106, -.35118919402718, -8.3518020608948, 10.200525036608]).reshape(7, 7) cov_colnames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] infocrit = np.array([ 33, np.nan, -52.969418473462, 7, 119.93883694692, 130.41438987719]) infocrit_colnames = 'N ll0 ll df AIC BIC'.split() infocrit_rownames = '.'.split() predicted = np.array([ 34.815238952637, .07249507308006, 7.3026847839355, .17909966409206, 1.2540435791016, .36725598573685, 3.9734709262848, .1719862818718, 2.0739872455597, .27532628178596, 1.1471545696259, .51580721139908, 1.7763512134552, .23559851944447, 2.2698366641998, .21655206382275, 1.6349502801895, .27835717797279, 2.7504913806915, .44458091259003, 2.862185716629, .54439353942871, 3.5617923736572, .57089400291443, 2.6135795116425, .41426089406013, .775799036026, .35101860761642, .93375068902969, .39217269420624, .56681954860687, .27232182025909, 1.8914022445679, .24083258211613]).reshape(17, 2) predicted_colnames = 'predict_mu predict_linpred_std'.split() predicted_rownames = ['r'+str(n) for n in range(1, 18)] results_poisson_fweight_clu1 = ParamsTableTestBunch( params_table=params_table, params_table_colnames=params_table_colnames, params_table_rownames=params_table_rownames, cov=cov, cov_colnames=cov_colnames, cov_rownames=cov_rownames, infocrit=infocrit, infocrit_colnames=infocrit_colnames, infocrit_rownames=infocrit_rownames, predicted=predicted, predicted_colnames=predicted_colnames, predicted_rownames=predicted_rownames, **est ) est = dict( k_eq_model=0, vf=1, df=10, df_m=6, power=0, canonical=1, rank=7, aic=4.033797198035106, rc=0, p=6.87057569091e-91, chi2=435.3803627057688, ll=-27.28727618329841, N_clust=9, k_autoCns=0, converged=1, k_dv=1, k_eq=1, k=7, ic=3, N=17, nbml=0, bic=-16.30350260869269, dispers_ps=1.548630027479802, deviance_ps=15.48630027479802, dispers_p=1.548630027479802, deviance_p=15.48630027479802, dispers_s=1.202863083186947, deviance_s=12.02863083186947, dispers=1.202863083186947, deviance=12.02863083186947, phi=1, cmdline="glm executions income perpoverty perblack LN_VC100k96 south degree [aweight=fweight], family(poisson) vce(cluster id)", # noqa:E501 cmd="glm", predict="glim_p", marginsnotok="stdp Anscombe Cooksd Deviance Hat Likelihood Pearson Response Score Working ADJusted STAndardized STUdentized MODified", # noqa:E501 marginsok="default", hac_lag="15", vcetype="Robust", vce="cluster", linkt="Log", linkf="ln(u)", varfunct="Poisson", varfuncf="u", opt1="ML", clustvar="id", oim="oim", a="1", m="1", varfunc="glim_v3", link="glim_l03", wexp="= fweight", wtype="aweight", chi2type="Wald", opt="moptimize", title="Generalized linear models", user="glim_lf", crittype="log pseudolikelihood", ml_method="e2", singularHmethod="m-marquardt", technique="nr", which="max", depvar="executions", properties="b V", ) params_table = np.array([ .00025343868829, .0000293670276, 8.6300422274633, 6.132932700e-18, .00019588037186, .00031099700472, np.nan, 1.9599639845401, 0, .09081422305585, .09800194027665, .92665739881771, .35410444288803, -.10126605030143, .28289449641312, np.nan, 1.9599639845401, 0, -.09416451429381, .02511206083893, -3.7497724658192, .00017699509401, -.14338324911569, -.04494577947192, np.nan, 1.9599639845401, 0, .27652273809507, .36749499887001, .75245306451881, .45177864537677, -.44375422418873, .99679970037887, np.nan, 1.9599639845401, 0, 2.239890838384, .51564197481271, 4.343887712395, .00001399830855, 1.229251138834, 3.250530537934, np.nan, 1.9599639845401, 0, -18.842583191417, 3.2292740757119, -5.8349284543965, 5.381365332e-09, -25.171844076021, -12.513322306812, np.nan, 1.9599639845401, 0, -6.5630017977417, 3.193826081147, -2.054902687558, .03988840483718, -12.822785889674, -.30321770580895, np.nan, 1.9599639845401, 0]).reshape(7, 9) params_table_colnames = 'b se z pvalue ll ul df crit eform'.split() params_table_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov = np.array([ 8.624223101e-10, 2.413510691e-06, 3.123995891e-07, -4.358439015e-06, -8.084672085e-06, -4.785328653e-06, -.00003652286809, 2.413510691e-06, .00960438029799, .00106422375754, -.00911884619892, -.03121758372723, .06803953530995, -.1771575604842, 3.123995891e-07, .00106422375754, .00063061559958, -.00844230553012, -.01177586448603, -.05361546061038, .03844868195581, -4.358439015e-06, -.00911884619892, -.00844230553012, .13505257419447, .1405885311093, .86184257188684, -.74146699290197, -8.084672085e-06, -.03121758372723, -.01177586448603, .1405885311093, .26588664618875, .75712244813928, -.35118919402768, -4.785328653e-06, .06803953530995, -.05361546061038, .86184257188684, .75712244813928, 10.428211056065, -8.3518020609031, -.00003652286809, -.1771575604842, .03844868195581, -.74146699290197, -.35118919402768, -8.3518020609031, 10.200525036615]).reshape(7, 7) cov_colnames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] infocrit = np.array([ 17, np.nan, -27.287276183298, 7, 68.574552366597, 74.40704577499]) infocrit_colnames = 'N ll0 ll df AIC BIC'.split() infocrit_rownames = '.'.split() predicted = np.array([ 34.815238952637, .07249507308006, 7.3026847839355, .17909966409206, 1.2540435791016, .36725598573685, 3.9734709262848, .1719862818718, 2.0739872455597, .27532628178596, 1.1471545696259, .51580721139908, 1.7763512134552, .23559851944447, 2.2698366641998, .21655206382275, 1.6349502801895, .27835714817047, 2.7504913806915, .44458091259003, 2.862185716629, .54439353942871, 3.5617923736572, .57089400291443, 2.6135795116425, .41426089406013, .775799036026, .35101860761642, .93375068902969, .39217269420624, .56681954860687, .27232182025909, 1.8914022445679, .24083258211613]).reshape(17, 2) predicted_colnames = 'predict_mu predict_linpred_std'.split() predicted_rownames = ['r'+str(n) for n in range(1, 18)] results_poisson_aweight_clu1 = ParamsTableTestBunch( params_table=params_table, params_table_colnames=params_table_colnames, params_table_rownames=params_table_rownames, cov=cov, cov_colnames=cov_colnames, cov_rownames=cov_rownames, infocrit=infocrit, infocrit_colnames=infocrit_colnames, infocrit_rownames=infocrit_rownames, predicted=predicted, predicted_colnames=predicted_colnames, predicted_rownames=predicted_rownames, **est ) est = dict( k_eq_model=0, vf=1, df=10, df_m=6, power=0, canonical=1, rank=7, aic=7.055225702760191, rc=0, p=6.87057569032e-91, chi2=435.380362705941, ll=-52.96941847346162, N_clust=9, k_autoCns=0, converged=1, k_dv=1, k_eq=1, k=7, ic=3, N=17, nbml=0, bic=-4.982438296344967, dispers_ps=3.006164170990202, deviance_ps=30.06164170990202, dispers_p=3.006164170990202, deviance_p=30.06164170990202, dispers_s=2.33496951442172, deviance_s=23.34969514421719, dispers=2.33496951442172, deviance=23.34969514421719, phi=1, cmdline="glm executions income perpoverty perblack LN_VC100k96 south degree [pweight=fweight], family(poisson) vce(cluster id)", # noqa:E501 cmd="glm", predict="glim_p", marginsnotok="stdp Anscombe Cooksd Deviance Hat Likelihood Pearson Response Score Working ADJusted STAndardized STUdentized MODified", # noqa:E501 marginsok="default", hac_lag="15", vcetype="Robust", vce="cluster", linkt="Log", linkf="ln(u)", varfunct="Poisson", varfuncf="u", opt1="ML", clustvar="id", oim="oim", a="1", m="1", varfunc="glim_v3", link="glim_l03", wexp="= fweight", wtype="pweight", chi2type="Wald", opt="moptimize", title="Generalized linear models", user="glim_lf", crittype="log pseudolikelihood", ml_method="e2", singularHmethod="m-marquardt", technique="nr", which="max", depvar="executions", properties="b V", ) params_table = np.array([ .00025343868829, .0000293670276, 8.6300422274613, 6.132932700e-18, .00019588037186, .00031099700472, np.nan, 1.9599639845401, 0, .09081422305585, .09800194027664, .92665739881773, .35410444288802, -.10126605030142, .28289449641311, np.nan, 1.9599639845401, 0, -.09416451429381, .02511206083893, -3.7497724658197, .00017699509401, -.14338324911569, -.04494577947193, np.nan, 1.9599639845401, 0, .27652273809506, .36749499886987, .75245306451906, .45177864537662, -.44375422418847, .99679970037859, np.nan, 1.9599639845401, 0, 2.239890838384, .51564197481271, 4.343887712395, .00001399830855, 1.229251138834, 3.250530537934, np.nan, 1.9599639845401, 0, -18.842583191417, 3.2292740757113, -5.8349284543976, 5.381365332e-09, -25.17184407602, -12.513322306813, np.nan, 1.9599639845401, 0, -6.5630017977416, 3.1938260811459, -2.0549026875586, .03988840483712, -12.822785889672, -.30321770581092, np.nan, 1.9599639845401, 0]).reshape(7, 9) params_table_colnames = 'b se z pvalue ll ul df crit eform'.split() params_table_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov = np.array([ 8.624223101e-10, 2.413510691e-06, 3.123995891e-07, -4.358439015e-06, -8.084672085e-06, -4.785328653e-06, -.00003652286809, 2.413510691e-06, .00960438029799, .00106422375754, -.00911884619892, -.03121758372723, .06803953530989, -.17715756048416, 3.123995891e-07, .00106422375754, .00063061559958, -.00844230553011, -.01177586448603, -.05361546061036, .03844868195577, -4.358439015e-06, -.00911884619892, -.00844230553011, .13505257419436, .14058853110927, .86184257188631, -.74146699290106, -8.084672085e-06, -.03121758372723, -.01177586448603, .14058853110927, .26588664618875, .75712244813913, -.35118919402718, -4.785328653e-06, .06803953530989, -.05361546061036, .86184257188631, .75712244813913, 10.428211056061, -8.3518020608948, -.00003652286809, -.17715756048416, .03844868195577, -.74146699290106, -.35118919402718, -8.3518020608948, 10.200525036608]).reshape(7, 7) cov_colnames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] infocrit = np.array([ 17, np.nan, -52.969418473462, 7, 119.93883694692, 125.77133035532]) infocrit_colnames = 'N ll0 ll df AIC BIC'.split() infocrit_rownames = '.'.split() predicted = np.array([ 34.815238952637, .07249507308006, 7.3026847839355, .17909966409206, 1.2540435791016, .36725598573685, 3.9734709262848, .1719862818718, 2.0739872455597, .27532628178596, 1.1471545696259, .51580721139908, 1.7763512134552, .23559851944447, 2.2698366641998, .21655206382275, 1.6349502801895, .27835717797279, 2.7504913806915, .44458091259003, 2.862185716629, .54439353942871, 3.5617923736572, .57089400291443, 2.6135795116425, .41426089406013, .775799036026, .35101860761642, .93375068902969, .39217269420624, .56681954860687, .27232182025909, 1.8914022445679, .24083258211613]).reshape(17, 2) predicted_colnames = 'predict_mu predict_linpred_std'.split() predicted_rownames = ['r'+str(n) for n in range(1, 18)] results_poisson_pweight_clu1 = ParamsTableTestBunch( params_table=params_table, params_table_colnames=params_table_colnames, params_table_rownames=params_table_rownames, cov=cov, cov_colnames=cov_colnames, cov_rownames=cov_rownames, infocrit=infocrit, infocrit_colnames=infocrit_colnames, infocrit_rownames=infocrit_rownames, predicted=predicted, predicted_colnames=predicted_colnames, predicted_rownames=predicted_rownames, **est ) est = dict( rank=7, ll_0=-55.23556912834824, ll=-47.54122045581504, r2_a=.3528737432046668, rss=267.3132086911238, mss=393.6105745962962, rmse=5.17023412130557, r2=.5955460895029168, F=.7279778160729128, df_r=10, df_m=6, N=17, cmdline="regress executions income perpoverty perblack LN_VC100k96 south degree [aweight=fweight], vce(robust)", # noqa:E501 title="Linear regression", marginsok="XB default", vce="robust", depvar="executions", cmd="regress", properties="b V", predict="regres_p", model="ols", estat_cmd="regress_estat", wexp="= fweight", wtype="aweight", vcetype="Robust", ) params_table = np.array([ .00177624355887, .00100571734546, 1.7661458926668, .10782432028789, -.00046463433267, .0040171214504, 10, 2.2281388519863, 0, .70240571372092, .54986275700055, 1.2774200557835, .23031379083217, -.5227648584123, 1.9275762858541, 10, 2.2281388519863, 0, -.76566360596606, .46482124106144, -1.6472216377583, .13053265392051, -1.8013498724035, .27002266047141, 10, 2.2281388519863, 0, 5.7915855647065, 5.8518623033717, .98969956305525, .34566324660643, -7.2471761899099, 18.830347319323, 10, 2.2281388519863, 0, 13.018291494864, 7.3741002410906, 1.7654074489417, .10795348742173, -3.412227750751, 29.44881074048, 10, 2.2281388519863, 0, -140.99921608421, 84.973820309491, -1.6593253730463, .12803894207791, -330.33268651749, 48.334254349065, 10, 2.2281388519863, 0, -68.484290889814, 50.764306481463, -1.3490638528633, .20706938025917, -181.5942144553, 44.625632675673, 10, 2.2281388519863, 0]).reshape(7, 9) params_table_colnames = 'b se t pvalue ll ul df crit eform'.split() params_table_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov = np.array([ 1.011467379e-06, .00038778854684, -.00038909911416, .00356508765632, .0056952104088, -.07926157334067, -.04218673068644, .00038778854684, .30234905153625, -.10112236243026, .59175926747871, 1.4744074711876, -25.6203584288, -14.793319880623, -.00038909911416, -.10112236243026, .21605878614189, -2.3405630815795, -3.2257627901142, 31.66920792546, 20.934058595259, .00356508765632, .59175926747871, -2.3405630815795, 34.244292417623, 34.810403897967, -270.34292245471, -270.19382562804, .0056952104088, 1.4744074711876, -3.2257627901142, 34.810403897967, 54.377354365652, -414.2817137548, -324.24739845086, -.07926157334067, -25.6203584288, 31.66920792546, -270.34292245471, -414.2817137548, 7220.5501379896, 2907.4556071681, -.04218673068644, -14.793319880623, 20.934058595259, -270.19382562804, -324.24739845086, 2907.4556071681, 2577.0148125439]).reshape(7, 7) cov_colnames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] infocrit = np.array([ 17, -55.235569128348, -47.541220455815, 7, 109.08244091163, 114.91493432002]) infocrit_colnames = 'N ll0 ll df AIC BIC'.split() infocrit_rownames = '.'.split() predicted = np.array([ 23.018356323242, 11.030969619751, 7.6487560272217, 3.2376720905304, 1.3298480510712, 2.4579885005951, 6.7120413780212, 2.8951823711395, .90416890382767, 2.1985862255096, 1.9608836174011, 2.5452246665955, 4.6054129600525, 2.8738057613373, 2.9902882575989, 1.8505314588547, 1.4887162446976, 1.47836124897, 5.9044842720032, 4.8891386985779, 7.0818486213684, 4.6786789894104, 7.5460968017578, 5.5129766464233, 4.1125593185425, 2.3989260196686, -2.7979807853699, 3.8943622112274, -1.4647831916809, 2.8729522228241, -3.5234127044678, 3.7701880931854, 3.9779393672943, 1.9573417901993]).reshape(17, 2) predicted_colnames = 'predict_mu predict_std'.split() predicted_rownames = ['r'+str(n) for n in range(1, 18)] results_wls_aweight_robust = ParamsTableTestBunch( params_table=params_table, params_table_colnames=params_table_colnames, params_table_rownames=params_table_rownames, cov=cov, cov_colnames=cov_colnames, cov_rownames=cov_rownames, infocrit=infocrit, infocrit_colnames=infocrit_colnames, infocrit_rownames=infocrit_rownames, predicted=predicted, predicted_colnames=predicted_colnames, predicted_rownames=predicted_rownames, **est ) est = dict( rank=7, ll_0=-55.23556912834824, ll=-47.54122045581504, r2_a=.3528737432046668, rss=267.3132086911238, mss=393.6105745962962, rmse=5.17023412130557, r2=.5955460895029168, F=1.412187242235973, df_r=8, df_m=6, N=17, N_clust=9, cmdline="regress executions income perpoverty perblack LN_VC100k96 south degree [aweight=fweight], vce(cluster id)", # noqa:E501 title="Linear regression", marginsok="XB default", vce="cluster", depvar="executions", cmd="regress", properties="b V", predict="regres_p", model="ols", estat_cmd="regress_estat", wexp="= fweight", wtype="aweight", vcetype="Robust", clustvar="id", ) params_table = np.array([ .00177624355887, .00103574504038, 1.7149428571794, .12469817836724, -.00061218878728, .00416467590501, 8, 2.3060041352042, 0, .70240571372092, .64463869959516, 1.0896114585768, .30761438040884, -.78413379325815, 2.1889452207, 8, 2.3060041352042, 0, -.76566360596606, .50850811868177, -1.5057057652313, .17056206446331, -1.9382854304311, .40695821849901, 8, 2.3060041352042, 0, 5.7915855647065, 6.2948340440059, .92005373362009, .3844480847801, -8.7243277711951, 20.307498900608, 8, 2.3060041352042, 0, 13.018291494864, 7.9526248350517, 1.6369804642972, .14027059672576, -5.3204942604922, 31.357077250221, 8, 2.3060041352042, 0, -140.99921608421, 84.897180497105, -1.6608233071889, .13532738016362, -336.77246537771, 54.774033209288, 8, 2.3060041352042, 0, -68.484290889814, 50.203382265366, -1.3641369923608, .2096627597382, -184.25349799498, 47.284916215355, 8, 2.3060041352042, 0]).reshape(7, 9) params_table_colnames = 'b se t pvalue ll ul df crit eform'.split() params_table_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov = np.array([ 1.072767789e-06, .00042569049255, -.00044272344175, .00386796354086, .00653558563917, -.08376884119522, -.04513384476642, .00042569049255, .41555905301573, -.07730648264729, -.34087330734824, .82631440946934, -31.768811666606, -10.324414524804, -.00044272344175, -.07730648264729, .25858050676528, -2.8727606144729, -3.9481543148554, 35.836754991381, 24.653552354067, .00386796354086, -.34087330734824, -2.8727606144729, 39.624935641576, 42.351437415382, -335.98208369348, -283.16728769825, .00653558563917, .82631440946934, -3.9481543148554, 42.351437415382, 63.24424176708, -502.21726015398, -366.49477518415, -.08376884119522, -31.768811666606, 35.836754991381, -335.98208369348, -502.21726015398, 7207.531256358, 3532.1379707168, -.04513384476642, -10.324414524804, 24.653552354067, -283.16728769825, -366.49477518415, 3532.1379707168, 2520.3795908825]).reshape(7, 7) cov_colnames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] infocrit = np.array([ 17, -55.235569128348, -47.541220455815, 7, 109.08244091163, 114.91493432002]) infocrit_colnames = 'N ll0 ll df AIC BIC'.split() infocrit_rownames = '.'.split() predicted = np.array([ 23.018356323242, 11.727355003357, 7.6487560272217, 3.4638004302979, 1.3298480510712, 2.1195623874664, 6.7120413780212, 2.8227334022522, .90416890382767, 2.2036759853363, 1.9608836174011, 2.0707910060883, 4.6054129600525, 2.9022018909454, 2.9902882575989, 1.6939970254898, 1.4887162446976, 1.8477793931961, 5.9044842720032, 4.8752007484436, 7.0818486213684, 4.4365234375, 7.5460968017578, 5.6850047111511, 4.1125593185425, 2.7407164573669, -2.7979807853699, 3.9614858627319, -1.4647831916809, 2.4376966953278, -3.5234127044678, 3.5529434680939, 3.9779393672943, 1.7075037956238]).reshape(17, 2) predicted_colnames = 'predict_mu predict_std'.split() predicted_rownames = ['r'+str(n) for n in range(1, 18)] results_wls_aweight_clu1 = ParamsTableTestBunch( params_table=params_table, params_table_colnames=params_table_colnames, params_table_rownames=params_table_rownames, cov=cov, cov_colnames=cov_colnames, cov_rownames=cov_rownames, infocrit=infocrit, infocrit_colnames=infocrit_colnames, infocrit_rownames=infocrit_rownames, predicted=predicted, predicted_colnames=predicted_colnames, predicted_rownames=predicted_rownames, **est ) est = dict( rank=7, ll_0=-107.2219871314995, ll=-92.28589853187629, r2_a=.5022105716958969, rss=518.9021109886529, mss=764.067585981045, rmse=4.467412394167744, r2=.5955460895029162, F=1.835843414931295, df_r=8, df_m=6, N=33, N_clust=9, cmdline="regress executions income perpoverty perblack LN_VC100k96 south degree [fweight=fweight], vce(cluster id)", # noqa:E501 title="Linear regression", marginsok="XB default", vce="cluster", depvar="executions", cmd="regress", properties="b V", predict="regres_p", model="ols", estat_cmd="regress_estat", wexp="= fweight", wtype="fweight", vcetype="Robust", clustvar="id", ) params_table = np.array([ .00177624355887, .00090840849363, 1.9553357012053, .08627786102497, -.00031855018389, .00387103730162, 8, 2.3060041352042, 0, .70240571372091, .56538554103558, 1.2423482079757, .24928937729829, -.60137568189177, 2.0061871093336, 8, 2.3060041352042, 0, -.76566360596606, .44599112337258, -1.7167687109468, .12435346910262, -1.7941209807276, .26279376879547, 8, 2.3060041352042, 0, 5.7915855647065, 5.5209346785031, 1.0490226568442, .32482245151877, -6.9397126341137, 18.522883763527, 8, 2.3060041352042, 0, 13.018291494864, 6.9749133861223, 1.866444896759, .09894610636006, -3.0658876162246, 29.102470605953, 8, 2.3060041352042, 0, -140.99921608421, 74.459752971542, -1.8936299202886, .09489418422765, -312.70371434287, 30.705282174445, 8, 2.3060041352042, 0, -68.484290889814, 44.031279012175, -1.5553554751584, .15847103736706, -170.02060237022, 33.05202059059, 8, 2.3060041352042, 0]).reshape(7, 9) params_table_colnames = 'b se t pvalue ll ul df crit eform'.split() params_table_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov = np.array([ 8.252059913e-07, .00032745422504, -.00034055649365, .00297535656989, .0050273735686, -.06443757015017, -.03471834212801, .00032745422504, .31966081001209, -.05946652511329, -.26221023642171, .63562646882257, -24.437547435849, -7.9418573267692, -.00034055649365, -.05946652511329, .19890808212714, -2.2098158572872, -3.037041780658, 27.566734608754, 18.96427104159, .00297535656989, -.26221023642171, -2.2098158572872, 30.480719724298, 32.578028781062, -258.44775668729, -217.82099053713, .0050273735686, .63562646882257, -3.037041780658, 32.578028781062, 48.649416743908, -386.32096934921, -281.91905783396, -.06443757015017, -24.437547435849, 27.566734608754, -258.44775668729, -386.32096934921, 5544.254812583, 2717.0292082435, -.03471834212801, -7.9418573267692, 18.96427104159, -217.82099053713, -281.91905783396, 2717.0292082435, 1938.753531448]).reshape(7, 7) cov_colnames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] infocrit = np.array([ 33, -107.2219871315, -92.285898531876, 7, 198.57179706375, 209.04734999402]) infocrit_colnames = 'N ll0 ll df AIC BIC'.split() infocrit_rownames = '.'.split() predicted = np.array([ 23.018356323242, 10.285571098328, 7.6487560272217, 3.0379540920258, 1.3298480510712, 1.8589791059494, 6.7120413780212, 2.4757008552551, .90416890382767, 1.9327516555786, 1.9608836174011, 1.8162038326263, 4.6054129600525, 2.5453994274139, 2.9902882575989, 1.485733628273, 1.4887162446976, 1.6206097602844, 5.9044842720032, 4.2758340835571, 7.0818486213684, 3.8910882472992, 7.5460968017578, 4.9860787391663, 4.1125593185425, 2.4037673473358, -2.7979807853699, 3.4744529724121, -1.4647831916809, 2.1380014419556, -3.5234127044678, 3.1161375045776, 3.9779393672943, 1.4975799322128]).reshape(17, 2) predicted_colnames = 'predict_mu predict_std'.split() predicted_rownames = ['r'+str(n) for n in range(1, 18)] results_wls_fweight_clu1 = ParamsTableTestBunch( params_table=params_table, params_table_colnames=params_table_colnames, params_table_rownames=params_table_rownames, cov=cov, cov_colnames=cov_colnames, cov_rownames=cov_rownames, infocrit=infocrit, infocrit_colnames=infocrit_colnames, infocrit_rownames=infocrit_rownames, predicted=predicted, predicted_colnames=predicted_colnames, predicted_rownames=predicted_rownames, **est ) est = dict( rank=7, ll_0=-55.23556912834824, ll=-47.54122045581504, r2_a=.3528737432046668, rss=267.3132086911238, mss=393.6105745962962, rmse=5.17023412130557, r2=.5955460895029168, F=1.412187242235973, df_r=8, df_m=6, N=17, N_clust=9, cmdline="regress executions income perpoverty perblack LN_VC100k96 south degree [pweight=fweight], vce(cluster id)", # noqa:E501 title="Linear regression", marginsok="XB default", vce="cluster", depvar="executions", cmd="regress", properties="b V", predict="regres_p", model="ols", estat_cmd="regress_estat", wexp="= fweight", wtype="pweight", vcetype="Robust", clustvar="id", ) params_table = np.array([ .00177624355887, .00103574504038, 1.7149428571794, .12469817836724, -.00061218878728, .00416467590501, 8, 2.3060041352042, 0, .70240571372092, .64463869959516, 1.0896114585768, .30761438040884, -.78413379325815, 2.1889452207, 8, 2.3060041352042, 0, -.76566360596606, .50850811868177, -1.5057057652313, .17056206446331, -1.9382854304311, .40695821849901, 8, 2.3060041352042, 0, 5.7915855647065, 6.2948340440059, .92005373362009, .3844480847801, -8.7243277711951, 20.307498900608, 8, 2.3060041352042, 0, 13.018291494864, 7.9526248350517, 1.6369804642972, .14027059672576, -5.3204942604922, 31.357077250221, 8, 2.3060041352042, 0, -140.99921608421, 84.897180497105, -1.6608233071889, .13532738016362, -336.77246537771, 54.774033209288, 8, 2.3060041352042, 0, -68.484290889814, 50.203382265366, -1.3641369923608, .2096627597382, -184.25349799498, 47.284916215355, 8, 2.3060041352042, 0]).reshape(7, 9) params_table_colnames = 'b se t pvalue ll ul df crit eform'.split() params_table_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov = np.array([ 1.072767789e-06, .00042569049255, -.00044272344175, .00386796354086, .00653558563917, -.08376884119522, -.04513384476642, .00042569049255, .41555905301573, -.07730648264729, -.34087330734824, .82631440946934, -31.768811666606, -10.324414524804, -.00044272344175, -.07730648264729, .25858050676528, -2.8727606144729, -3.9481543148554, 35.836754991381, 24.653552354067, .00386796354086, -.34087330734824, -2.8727606144729, 39.624935641576, 42.351437415382, -335.98208369348, -283.16728769825, .00653558563917, .82631440946934, -3.9481543148554, 42.351437415382, 63.24424176708, -502.21726015398, -366.49477518415, -.08376884119522, -31.768811666606, 35.836754991381, -335.98208369348, -502.21726015398, 7207.531256358, 3532.1379707168, -.04513384476642, -10.324414524804, 24.653552354067, -283.16728769825, -366.49477518415, 3532.1379707168, 2520.3795908825]).reshape(7, 7) cov_colnames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] cov_rownames = ['income', 'perpoverty', 'perblack', 'LN_VC100k96', 'south', 'degree', '_cons'] infocrit = np.array([ 17, -55.235569128348, -47.541220455815, 7, 109.08244091163, 114.91493432002]) infocrit_colnames = 'N ll0 ll df AIC BIC'.split() infocrit_rownames = '.'.split() predicted = np.array([ 23.018356323242, 11.727355003357, 7.6487560272217, 3.4638004302979, 1.3298480510712, 2.1195623874664, 6.7120413780212, 2.8227334022522, .90416890382767, 2.2036759853363, 1.9608836174011, 2.0707910060883, 4.6054129600525, 2.9022018909454, 2.9902882575989, 1.6939970254898, 1.4887162446976, 1.8477793931961, 5.9044842720032, 4.8752007484436, 7.0818486213684, 4.4365234375, 7.5460968017578, 5.6850047111511, 4.1125593185425, 2.7407164573669, -2.7979807853699, 3.9614858627319, -1.4647831916809, 2.4376966953278, -3.5234127044678, 3.5529434680939, 3.9779393672943, 1.7075037956238]).reshape(17, 2) predicted_colnames = 'predict_mu predict_std'.split() predicted_rownames = ['r'+str(n) for n in range(1, 18)] results_wls_pweight_clu1 = ParamsTableTestBunch( params_table=params_table, params_table_colnames=params_table_colnames, params_table_rownames=params_table_rownames, cov=cov, cov_colnames=cov_colnames, cov_rownames=cov_rownames, infocrit=infocrit, infocrit_colnames=infocrit_colnames, infocrit_rownames=infocrit_rownames, predicted=predicted, predicted_colnames=predicted_colnames, predicted_rownames=predicted_rownames, **est )