M/PhIXdZddlZddlmcmZddlmcmZ ddl mcm Z ddl mZmZddlZddlZddlZGddejZGddeZGdd eZGd d ejZGd d eZGdde jZe jeedS)zA Conditional logistic, Poisson, and multinomial logit regression N)MultinomialResultsMultinomialResultsWrappercteZdZdfd ZdZ dfd Z dd Ze dfd ZxZ S)_ConditionalModelnonec d|vrtd|d}|j|jkrd}t||jd|jkrd}t|tj||fd|i||jjd}t||j}|jd|_i}t|D])\}} | |vrg|| <||  |*tj |tj |}}| d } g|_g|_g|_| tj | } g|_g|_g|_d|_ddg} |D])\} } || j} tj| dkr.| dxxdz cc<| dxxt1| z cc<Y|xjt1| z c_|j | | |j | | |j t1| |j || ddf|j tj| +| ddkr&d t5| z}t7j|| Yg|_t|jD]=\}}|j tj|j||>t1|j|_g|_ g|_!tE|jD]|} |j tj|j| |j| |j! tj|j| }dS) Ngroups'groups' is a required argumentz4'endog' and 'groups' should have the same dimensionsrzBThe leading dimension of 'exog' should equal the length of 'endog'missingzDConditional models should not have an intercept in the design matrixoffsetzIDropped %d groups and %d observations for having no within-group variance)# ValueErrorsizeshapesuper__init__data const_idxexogk_params enumerateappendnpasarrayget _endog_grp _exog_grp _groupsize _offset_grp_offset_sumynobsitemsflatstdlensumtuplewarningswarn_endofsdot _n_groups_xy_n1range)selfendogrr kwargsr msgrow_ixigr dropsixykofs __class__s g/var/www/html/test/jupyter/venv/lib/python3.11/site-packages/statsmodels/discrete/conditional_models.pyrz_ConditionalModel.__init__s 6 ! !>?? ?! ;%* $ $HCS// ! :a=EJ & &VCS// ! 4 4 4!( 4,2 4 4 4 9  *"CS// !y 1  f%%  DAqq 1I  Q    j''D)9)9tH%%  Z''F!D    A\\^^ ) )EArb AvayyA~~aA aCFF" IIQ II O " "1 % % %! ''r 333 O " "3q66 * * * N ! !$r111u+ . . . J  bfQii ( ( ( ( 8a<<.16u>C M#     DL#D$455 E E3 ##BF4?1+=s$C$CDDDDT_--t~&& 8 8A HOOBF4?1#5t~a7HII J J J HOOBF4?1#566 7 7 7 7 8 8c\ddlm}|||j}tj|}|S)Nr) approx_fprime)statsmodels.tools.numdiffrAscorer atleast_2d)r1paramsrAhesss r>hessianz_ConditionalModel.hessianbs;;;;;;;}VTZ00}T"" r?NBFGSdTFc  t|||||| } t|| j| d} || _|j| _|j| _dt|j zdt|j zdtj |j zg| _t| } | S)N start_paramsmethodmaxiter full_outputdisp skip_hessianr z%dz%.1f)rfitConditionalResultsrE cov_paramsrNr"r-n_groupsminrmaxrmean _group_statsConditionalResultsWrapper)r1rMrNrOrPrQfargscallbackretallrRr3rsltcrsltr=s r>rSz_ConditionalModel.fitisww{{%#% ''#4doo6G6GKK Y  3t'' ' 3t'' ' RWT_-- -  )// r? elastic_netc ddlm}|dkrtdddddd}||||f||||d |S) a Return a regularized fit to a linear regression model. Parameters ---------- method : {'elastic_net'} Only the `elastic_net` approach is currently implemented. alpha : scalar or array_like The penalty weight. If a scalar, the same penalty weight applies to all variables in the model. If a vector, it must have the same length as `params`, and contains a penalty weight for each coefficient. start_params : array_like Starting values for `params`. refit : bool If True, the model is refit using only the variables that have non-zero coefficients in the regularized fit. The refitted model is not regularized. **kwargs Additional keyword argument that are used when fitting the model. Returns ------- Results A results instance. r)fit_elasticnetraz.method for fit_regularized must be elastic_net2r g|=)rOL1_wt cnvrg_tolzero_tol)rNalpharMrefit)statsmodels.base.elastic_netrdrupdate)r1rNrirMrjr3rddefaultss r>fit_regularizedz!_ConditionalModel.fit_regularizedsB @????? ] " "MNN N!AE %''~d*6$)+7$)**!) ** *r?c. |d}|d=n#t$rtdwxYwt|tr||}d|ddvrt jdtj|g|R||d|}|S)Nr r z0+ z2Conditional models should not include an intercept)rr ) KeyErrorr isinstancestrreplacer)r*r from_formula) clsformularsubset drop_colsargsr3r modelr=s r>rvz_ConditionalModel.from_formulas @H%Fx   @ @ @>?? ? @ fc " " "&\F wsB// / / MN O O O$$ @04@@v@@8>@@ s )r NrHrITFrJNFF)rarbNF)NN) __name__ __module__ __qualname__rrGrSrn classmethodrv __classcell__r=s@r>rrsN8N8N8N8N8N8`B - %)# .*.*.*.*b!# [r?rcLeZdZdZd fd ZdZdZd dZd dZd Z d Z xZ S) ConditionalLogita Fit a conditional logistic regression model to grouped data. Every group is implicitly given an intercept, but the model is fit using a conditional likelihood in which the intercepts are not present. Thus, intercept estimates are not given, but the other parameter estimates can be interpreted as being adjusted for any group-level confounders. Parameters ---------- endog : array_like The response variable, must contain only 0 and 1. exog : array_like The array of covariates. Do not include an intercept in this array. groups : array_like Codes defining the groups. This is a required keyword parameter. rc tj||fd|i|tjtj|jtjdkrd}t||jj d|_ dS)Nr )rr zendog must be coded as 0, 1r ) rrranyuniquer2r_rrrK)r1r2rr r3r4r=s r>rzConditionalLogit.__init__s 4 4 4!( 4,2 4 4 4 6")DJ''25;6 7 7 "/CS// !#r?cd}tt|jD]}||||z }|SNr)r0r&r loglike_grp)r1rEllr7s r>loglikezConditionalLogit.loglikesI s4?++,, . .A $""1f-- -BB r?cjd}t|jD]}||||z }|Sr)r0r- score_grp)r1rErCr7s r>rCzConditionalLogit.scores?t~&& / /A T^^Av.. .EE r?Nc|d}tjtj|j|||zifd|j||j|S)Nrc||krdS|dkrdS ||fS#t$rYnwxYw|dz ||dz |dz |dz zz}|||f<|S)Nrr )rr)tr;vexbfmemos r>rz"ConditionalLogit._denom..fs1uuqAvvq QF|#    !a% aaAq1uooAE ::AD!QLHs  **)rexpr,rrr/)r1grprEr<rrrs @@@r>_denomzConditionalLogit._denoms ;CfRVDN3/883>??        q%tx}555r?c|d}j|tjtj||zifdj|j|S)NrcZ||krdtjjfS|dkrdS ||fS#t$rYnwxYw |dz } |dz |\}} |dz |dz \}}||z |dz ddfz}|||zz||z||zz} }|| f ||f<|| fS)Nr)r rr )rzerosrrr)rr;habcedurexrrsr1s r>rz'ConditionalLogit._denom_grad..s.s1uu"(4=1111Avvt QF|#    AE A1QUA;;DAq1QUAE??DAqA1q5!!!8 $Aq1u9a!ea!emqAq6D!QLa4Ks 6 AA)rrrr,rr/)r1rrEr<rrrrs` @@@@r> _denom_gradzConditionalLogit._denom_grad"s ;C ^C fRVB''#-..         ,q%tx}555r?cd}t|dr |j|}tj|j||}|||j|z }|tj||||z}|S)Nr )hasattrrrr,r.r+logr)r1rrEr<llgs r>rzConditionalLogit.loglike_grpFs| 4 " " ("3'CfTXc]F++ ? 4<$ $C rvdkk#vs33444 r?cd}t|dr |j|}||||\}}|j|||z z S)Nrr )rrrr.)r1rrEr<rrs r>rzConditionalLogit.score_grpUsV 4 " " ("3'CVS111x}q1u$$r?r})N) rrr__doc__rrrCrrrrrrs@r>rrs& $ $ $ $ $ $6666:"6"6"6"6H   %%%%%%%r?rceZdZdZdZdZdS)ConditionalPoissonaU Fit a conditional Poisson regression model to grouped data. Every group is implicitly given an intercept, but the model is fit using a conditional likelihood in which the intercepts are not present. Thus, intercept estimates are not given, but the other parameter estimates can be interpreted as being adjusted for any group-level confounders. Parameters ---------- endog : array_like The response variable exog : array_like The covariates groups : array_like Codes defining the groups. This is a required keyword parameter. cd}t|dr|j}d}tt|jD]}t j|j||}| |||z }t j|}|j|}|t j||z }| }||j |t j |zz}|SNr rb) rrr0r&rrr,rrr'r!r) r1rEr<rr6xbrr:rs r>rzConditionalPoisson.loglikers 4 " " #"C s4?++,, , ,Aq)622Bc!f &**C"A "&B-- B A $*Q-"&))+ +BB r?cd}t|dr|j}d}tt|jD]}|j|}t j||}| |||z }t j|}| }|j|} |t j| |z }||j |t j||z|z z}|Sr) rrr0r&rrrr,rr'r!) r1rEr<rCr6xrrrr:s r>rCzConditionalPoisson.scores 4 " " #"Cs4?++,, 8 8Aq!A6""Bc!f &**C A"A RVAq\\ !E TZ]RVC^^3a7 7EE r?N)rrrrrrCrJr?r>rr_s<$*r?rc&eZdZfdZddZxZS)rTcRt||||dS)N)normalized_cov_paramsscale)rr)r1r|rErrr=s r>rzConditionalResults.__init__s:   "7      r?N皙?c4dddd|jgfddg}dd|jgfd |jd gfd |jd gfd |jdgfg}|d}d dlm}|}|||||||||||||j|S)a< Summarize the fitted model. Parameters ---------- yname : str, optional Default is `y` xname : list[str], optional Names for the exogenous variables, default is "var_xx". Must match the number of parameters in the model title : str, optional Title for the top table. If not None, then this replaces the default title alpha : float Significance level for the confidence intervals Returns ------- smry : Summary instance This holds the summary tables and text, which can be printed or converted to various output formats. See Also -------- statsmodels.iolib.summary.Summary : class to hold summary results )zDep. Variable:N)zModel:N)zLog-Likelihood:NzMethod:)zDate:N)zTime:N)zNo. Observations:Nz No. groups:zMin group size:rzMax group size:r zMean group size:Nz*Conditional Logit Model Regression Results)Summary)gleftgrightynamexnametitle)rrriuse_t)rNrVrZstatsmodels.iolib.summaryradd_table_2colsadd_table_paramsr) r1rrrritop_left top_rightrsmrys r>summaryzConditionalResults.summarys< %  %  &     ( T]O , !21!5 6 7 !21!5 6 7 $"3A"6!7 8   =@E 655555wyy        U%tz  K K K r?)NNNr)rrrrrrrs@r>rTrTsL>>>>>>>>r?rTcJeZdZdZd fd Z dd Zd Zd ZxZS)ConditionalMNLogita Fit a conditional multinomial logit model to grouped data. Parameters ---------- endog : array_like The dependent variable, must be integer-valued, coded 0, 1, ..., c-1, where c is the number of response categories. exog : array_like The independent variables. groups : array_like Codes defining the groups. This is a required keyword parameter. Notes ----- Equivalent to femlogit in Stata. References ---------- Gary Chamberlain (1980). Analysis of covariance with qualitative data. The Review of Economic Studies. Vol. 47, No. 1, pp. 225-238. rc ltj||fd|i||jt|_|jdz|_|jdz |jjdz|_ |j |j z |_ dt|jD|_ |j|_|jjd|_|jdkrd}t#|t%jt(t+|jD] \}}||!t)|_|jfd|jD|_dS)Nr r c.i|]}|t|SrJ)rt).0js r> z/ConditionalMNLogit.__init__.. s AAA!As1vvAAAr?rz%endog may not contain negative valuesc g|] }| SrJrJ)rr;grxs r> z/ConditionalMNLogit.__init__..s;;;1A;;;r?)rrr2astypeintrXk_catrrdf_modelr"df_residr0 _ynames_mapJrrWr collections defaultdictlistrr rkeys _group_labelssort_grp_ix) r1r2rr r3r4r;rrr=s @r>rzConditionalMNLogit.__init__s 4 4 4!( 4,2 4 4 4Z&&s++ Z^^%%) a49?1+==  DM1 AAuTZ/@/@AAA# :>>  a  9CS// !%d++dk**  DAq FMM!    !#((**-- !!!;;;;(:;;; r?NrHrITFrJc |?|jjd} |jdz } tj| | z}t j||||||| } | j |jjddf| _ t|| } | tj t| S)Nr )rrL)llnull)rrrrrandomnormalbaseLikelihoodModelrSrEreshaperset_null_optionsnanr)r1rMrNrOrPrQr\r]r^rRr3qrr_s r>rSzConditionalMNLogit.fits   "A QA9++Q+77L#'' %#%(''k))49?1+=r*BCC !$-- RV,,,(...r?c|jjd}|jdz }|||f}t jt j|df|fd}t j|j|}d}|jD]}||ddf}t j |jdt} |j |} d} tj | D]3} | t j|| | fz } 4||| | ft j| z z }|S)Nr axisrbrdtype)rrrrr concatenaterr,rarangerr2 itertools permutationsrr'r) r1rErrpmatlprriirjjr:denomps r>rzConditionalMNLogit.loglike?s- IOA  JN~~q!f%%~rxA//6Q???fTY%% , 3 3BBE A171:S111B 2AE+A.. 2 22q'  0 0111 !RG*..""RVE]]2 2BB r?c |jjd}|jdz }|||f}t jt j|df|fd}t j|j|}t j||f}|jD]=}||ddf}t j |jdt} |j |} d} t j||f} tj | D]} t j|| | f}| |z } t!| D]:\}}|dkr/| dd|dz fxx||j||ddfzz cc<;t!| D]7\}}|dkr,|dd|dz fxx|j||ddfz cc<8|| | z z}?|S)Nr rrrrb)rrrrrrrr,rrrr2rrrr'rflatten)r1rErrrrgradrrrr:rdenomgrrr6rs r>rCzConditionalMNLogit.scoreUs IOA  JN~~q!f%%~rxA//6Q???fTY%%xA, # #BBE A171:S111B 2AEXq!f%%F+A.. D DF1b!W:>>++,, %aLLDDDAqAvvqqq!a%x(((A "Q%(0C,CC(((D"!  : :166AENNNdi1qqq&99NNN FUN "DD||~~r?r}r~) rrrrrrSrrCrrs@r>rrs0<<<<<<6#/#/#/#/J,r?rceZdZdS)r[N)rrrrJr?r>r[r[vsDr?r[)rnumpyrstatsmodels.base.modelrr|#statsmodels.regression.linear_model regression linear_modellmstatsmodels.base.wrapperwrapperwrap#statsmodels.discrete.discrete_modelrrrr)rrrrrLikelihoodModelResultsrTrRegressionResultsWrapperr[populate_wrapperrJr?r>rs%%%%%%%%%000000000'''''''''!!!!!!!!BBBBB,BBBJG%G%G%G%G%(G%G%G%T<<<<<*<<<~GGGGG4GGGRKKKKK*KKK^      ;   /1CDDDDDr?