M/PhɭdZddlmZddlmZmZddlZddlZ ddl m Z ddl m Z ddlmZddlmZdd lmZdd lmZmZmZdd lmZdd lmZd dlmZmZmZm Z m!Z!gdZ"dZ#d2dZ$d3dZ%d4dZ&dZ'dididdd fdZ(d5dZ)d4dZ*d5dZ+ d6dZ,eej-d7iddi d8d#Z.eej-d7idd$i d9d%Z/ee j-dd&id:d'Z0ee j-dd(id:d)Z1eedd(iz d;d*Z2ee!dd(izd4d+Z3eedd,iz dd1Z7dS)?zPartial Regression plot and residual plots to find misspecification Author: Josef Perktold License: BSD-3 Created: 2011-01-23 update 2011-06-05 : start to convert example to usable functions 2011-10-27 : docstrings )Appender)lrangelzipN)dmatrix)GEE)GLM)utils)lowess)GLSOLSWLS)wls_prediction_std)maybe_unwrap_results)_plot_added_variable_doc_plot_ceres_residuals_doc_plot_influence_doc_plot_leverage_resid2_doc_plot_partial_residuals_doc)plot_fitplot_regress_exogplot_partregress plot_ccprrplot_partregress_gridplot_ccpr_grid add_lowess abline_plotinfluence_plotplot_leverage_resid2added_variable_residspartial_resids ceres_residsplot_added_variableplot_partial_residualsplot_ceres_residualsc,d|jdzz|jz S)N@r)df_modelnobs)resultss d/var/www/html/test/jupyter/venv/lib/python3.11/site-packages/statsmodels/graphics/regressionplots.py_high_leverager,,s !A% &w| 33皙?c ||j}||j}t||fd|i|}||dddf|dddfdd|jS)a Add Lowess line to a plot. Parameters ---------- ax : AxesSubplot The Axes to which to add the plot lines_idx : int This is the line on the existing plot to which you want to add a smoothed lowess line. frac : float The fraction of the points to use when doing the lowess fit. lowess_kwargs Additional keyword arguments are passes to lowess. Returns ------- Figure The figure that holds the instance. fracNrrrg?)lw) get_lines_y_xr plotfigure)ax lines_idxr0 lowess_kwargsy0x0lress r+rr1s*  " %B  " %B "b 5 5t 5} 5 5DGGDAJQQQT CCG000 9r-Tc "tj|\}}tj||j\}}t |}|jj}|jjdd|f} tj| } || }| | } | | |d|jj || | || ddd|z} |j | |j | dfdd d ||d ur:t|\} } }| | | | || d d d|| ||||jj |dd |S)aQ Plot fit against one regressor. This creates one graph with the scatterplot of observed values compared to fitted values. Parameters ---------- results : Results A result instance with resid, model.endog and model.exog as attributes. exog_idx : {int, str} Name or index of regressor in exog matrix. y_true : array_like. optional If this is not None, then the array is added to the plot. ax : AxesSubplot, optional If given, this subplot is used to plot in instead of a new figure being created. vlines : bool, optional If this not True, then the uncertainty (pointwise prediction intervals) of the fit is not plotted. **kwargs The keyword arguments are passed to the plot command for the fitted values points. Returns ------- Figure If `ax` is None, the created figure. Otherwise the figure to which `ax` is connected. Examples -------- Load the Statewide Crime data set and perform linear regression with `poverty` and `hs_grad` as variables and `murder` as the response >>> import statsmodels.api as sm >>> import matplotlib.pyplot as plt >>> data = sm.datasets.statecrime.load_pandas().data >>> murder = data['murder'] >>> X = data[['poverty', 'hs_grad']] >>> X["constant"] = 1 >>> y = murder >>> model = sm.OLS(y, X) >>> results = model.fit() Create a plot just for the variable 'Poverty.' Note that vertical bars representing uncertainty are plotted since vlines is true >>> fig, ax = plt.subplots() >>> fig = sm.graphics.plot_fit(results, 0, ax=ax) >>> ax.set_ylabel("Murder Rate") >>> ax.set_xlabel("Poverty Level") >>> ax.set_title("Linear Regression") >>> plt.show() .. plot:: plots/graphics_plot_fit_ex.py Nbo)labelzb-z True valueszFitted values versus %sDr1fitted)colorr@Trkffffff? linewidthrCalphabest)loc numpoints)r create_mpl_axmaybe_name_or_idxmodelrendogexognpargsortr6 endog_names fittedvaluesrvlines set_title set_xlabel set_ylabellegend)r*exog_idxy_truer8rUkwargsfig exog_nameyx1 x1_argsorttitle_iv_liv_us r+rrMs~!"%%GC1(GMJJIx"7++G  A  AAAxK (BBJ * A JBGGB4w}8G999  F:&MBBB % 1E BGB$Z0#&S&&$&&& ~~*733 4 "d:&Z(8A2  ' ' 'LLMM)MM'-+,,,II&AI&&& Jr-c(tj|}tj||j\}}t |}|jj}|jjdd|f}t|\}}}|ddd} | ||jj ddd|| ||j dd d d | |||dd d| dd| || || d|ddd} | ||jd| dd| d|zd| || d|ddd} t'j|jjjdt,} d| |<|jjdd| f} ddlm} t3|jjj| |||jjj| d| }| dd|ddd } t;||| !}| d"d|d#|zd|| d$|S)%aPlot regression results against one regressor. This plots four graphs in a 2 by 2 figure: 'endog versus exog', 'residuals versus exog', 'fitted versus exog' and 'fitted plus residual versus exog' Parameters ---------- results : result instance A result instance with resid, model.endog and model.exog as attributes. exog_idx : int or str Name or index of regressor in exog matrix. fig : Figure, optional If given, this figure is simply returned. Otherwise a new figure is created. Returns ------- Figure The value of `fig` if provided. Otherwise a new instance. Examples -------- Load the Statewide Crime data set and build a model with regressors including the rate of high school graduation (hs_grad), population in urban areas (urban), households below poverty line (poverty), and single person households (single). Outcome variable is the murder rate (murder). Build a 2 by 2 figure based on poverty showing fitted versus actual murder rate, residuals versus the poverty rate, partial regression plot of poverty, and CCPR plot for poverty rate. >>> import statsmodels.api as sm >>> import matplotlib.pyplot as plt >>> import statsmodels.formula.api as smf >>> fig = plt.figure(figsize=(8, 6)) >>> crime_data = sm.datasets.statecrime.load_pandas() >>> results = smf.ols('murder ~ hs_grad + urban + poverty + single', ... data=crime_data.data).fit() >>> sm.graphics.plot_regress_exog(results, 'poverty', fig=fig) >>> plt.show() .. plot:: plots/graphics_regression_regress_exog.py Nrobg?)rCrHr@rAr1rB?)rCr@rHrDrErFzY and Fitted vs. XlargefontsizerI)rJrblack)r_rCzResiduals versus %sresidF)Series)nameindex) obs_labelsr8zPartial regression plotr8z CCPR PlotzRegression Plots for %stop)!r create_mpl_figrMrNrrSrPr add_subplotr6rOrTrUrVrWrXrYroaxhlinerQonesshapeboolpandasrqrdata orig_endog row_labelsrsuptitle tight_layoutsubplots_adjust) r*rZr]r^y_namer`prstdrdrer8 exog_noti exog_othersrqs r+rrs^  s # #C1(GMJJIx"7++G] &F  AAAxK (B*733E4 Aq ! !BGGB #S3fGMMMGGB$cH IIb$BI???LL%L888MM)MM&II&I Aq ! !BGGB s###JJ'J"""LL&2WLEEEMM)MM' Aq ! !B *03T::IIh-$QQQ \2K 7=-8!6"9(/ (:(EGGG&5R A A ACLL*WL=== Aq ! !B GX" - - -CLLwL///LL*Y6LIIIC   Jr-ct||}t||}t|j|j}|||ffS)aPartial regression. regress endog on exog_i conditional on exog_others uses OLS Parameters ---------- endog : array_like exog : array_like exog_others : array_like Returns ------- res1c : OLS results instance (res1a, res1b) : tuple of OLS results instances results from regression of endog on exog_others and of exog_i on exog_others )r fitro)rOexog_irres1ares1bres1cs r+_partial_regressionrsh, { # # ' ' ) )E  $ $ ( ( * *E  U[ ) ) - - / /E 5%.  r-Fc tj|\} }t|trt |dz|| }t|trt ||| } n?t|t r(d|} t | || } n|} d} t| tjr| j dkrd} n#t| tj r | j rd} t|trt |dz|| }| rtj |}tj |}|j||dfi| t||}t|tjrdn|j}t|tjrd n|jjd}nt|| }t|| }|j}|j}|jj}|jj}|j||dfi| t||}t/dtj |jdd | } |d krd}|d |z|d |z|jdi||durQ||j}n)t;|dr|j}n|jjj}|tAtC|}|durtC|tC|krtEd|#tIddtj%tAtC||tM|j|jdgtC|zdfd|i|}|r| |j|jffS| S)aN Plot partial regression for a single regressor. Parameters ---------- endog : {ndarray, str} The endogenous or response variable. If string is given, you can use a arbitrary translations as with a formula. exog_i : {ndarray, str} The exogenous, explanatory variable. If string is given, you can use a arbitrary translations as with a formula. exog_others : {ndarray, list[str]} Any other exogenous, explanatory variables. If a list of strings is given, each item is a term in formula. You can use a arbitrary translations as with a formula. The effect of these variables will be removed by OLS regression. data : {DataFrame, dict} Some kind of data structure with names if the other variables are given as strings. title_kwargs : dict Keyword arguments to pass on for the title. The key to control the fonts is fontdict. obs_labels : {bool, array_like} Whether or not to annotate the plot points with their observation labels. If obs_labels is a boolean, the point labels will try to do the right thing. First it will try to use the index of data, then fall back to the index of exog_i. Alternatively, you may give an array-like object corresponding to the observation numbers. label_kwargs : dict Keyword arguments that control annotate for the observation labels. ax : AxesSubplot, optional If given, this subplot is used to plot in instead of a new figure being created. ret_coords : bool If True will return the coordinates of the points in the plot. You can use this to add your own annotations. eval_env : int Patsy eval environment if user functions and formulas are used in defining endog or exog. **kwargs The keyword arguments passed to plot for the points. Returns ------- fig : Figure If `ax` is None, the created figure. Otherwise the figure to which `ax` is connected. coords : list, optional If ret_coords is True, return a tuple of arrays (x_coords, y_coords). See Also -------- plot_partregress_grid : Plot partial regression for a set of regressors. Notes ----- The slope of the fitted line is the that of `exog_i` in the full multiple regression. The individual points can be used to assess the influence of points on the estimated coefficient. Examples -------- Load the Statewide Crime data set and plot partial regression of the rate of high school graduation (hs_grad) on the murder rate(murder). The effects of the percent of the population living in urban areas (urban), below the poverty line (poverty) , and in a single person household (single) are removed by OLS regression. >>> import statsmodels.api as sm >>> import matplotlib.pyplot as plt >>> crime_data = sm.datasets.statecrime.load_pandas() >>> sm.graphics.plot_partregress(endog='murder', exog_i='hs_grad', ... exog_others=['urban', 'poverty', 'single'], ... data=crime_data.data, obs_labels=False) >>> plt.show() .. plot:: plots/graphics_regression_partregress.py More detailed examples can be found in the Regression Plots notebook on the examples page. z-1)eval_env+FrTrhxr_rD)rCr8z e(%s | X)Partial Regression PlotNrsz*obs_labels does not match length of exog_icenterbottom)havarx-larger8)r)'r rL isinstancestrrlistjoinrQndarraysizepd DataFrameemptyasarrayr6r rrr design_info column_namesrorNrSrparamsrWrXrVrshasattrrrrlen ValueErrorupdatedict annotate_axesr)rOrrr title_kwargsrt label_kwargsr8 ret_coordsrr\r]RHS RHS_isemtpy fitted_linex_axis_endog_namey_axis_endog_name res_yaxis res_xaxis xaxis_resid yaxis_resids r+rr/sn!"%%GC%? dX>>>+s##k4(;;; K & &hh{##c4(333K#rz""sx{{ C & &39 &#A$x@@@: 5!!F##vs--f---%((,,.. #-fbj#A#ARCCv{#-eRZ#@#@gCCeFWFdefFgsOO'')) $$((** o o %O7%O7 [#88888+{337799 aK$677:#" M M MCCMM+ 11222MM+ 11222BL;;l;;;T  JJ VW % % 9JJ"-8J  F ,,J z??c&kk ) )IJJ JDH:::;;;  J!8!8*!%ioy!G!G"(C OO!;Y11KM1$011 Y_io666 r-c ddl}tj|}tj||j\}}||jj|jj}|jj}|j d}t|dzdz} | t|krdnd} ||\} } | dkrdddii} tj |jj } t|D]\} }t|}||||dd|f| | }|| | | dz}t'|||dd|f| |||| d |d |d d||d|S)a Plot partial regression for a set of regressors. Parameters ---------- results : Results instance A regression model results instance. exog_idx : {None, list[int], list[str]} The indices or column names of the exog used in the plot, default is all. grid : {None, tuple[int]} If grid is given, then it is used for the arrangement of the subplots. The format of grid is (nrows, ncols). If grid is None, then ncol is one, if there are only 2 subplots, and the number of columns is two otherwise. fig : Figure, optional If given, this figure is simply returned. Otherwise a new figure is created. Returns ------- Figure If `fig` is None, the created figure. Otherwise `fig` itself. See Also -------- plot_partregress : Plot partial regression for a single regressor. plot_ccpr : Plot CCPR against one regressor Notes ----- A subplot is created for each explanatory variable given by exog_idx. The partial regression plot shows the relationship between the response and the given explanatory variable after removing the effect of all other explanatory variables in exog. References ---------- See http://www.itl.nist.gov/div898/software/dataplot/refman1/auxillar/partregr.htm Examples -------- Using the state crime dataset separately plot the effect of the each variable on the on the outcome, murder rate while accounting for the effect of all other variables in the model visualized with a grid of partial regression plots. >>> from statsmodels.graphics.regressionplots import plot_partregress_grid >>> import statsmodels.api as sm >>> import matplotlib.pyplot as plt >>> import statsmodels.formula.api as smf >>> fig = plt.figure(figsize=(8, 6)) >>> crime_data = sm.datasets.statecrime.load_pandas() >>> results = smf.ols('murder ~ hs_grad + urban + poverty + single', ... data=crime_data.data).fit() >>> plot_partregress_grid(results, fig=fig) >>> plt.show() .. plot:: plots/graphics_regression_partregress_grid.py rN)rrrrgfontdictrmsmall)columnsF)r8rrtrrkrlffffff?rw)rr ryrMrNrqrOrSrPr}rrQarray exog_names enumeraterpoprrzrrVrrr)r*rZgridr]rr^r_rPk_varsnrowsncolsr other_namesiidxothersrr8s r+rrs|MMM  s # #C1(GMJJIx  gm) 0I JJA = D Z]F]]Q 1 $E#h--''AAQE  u qyy"Z$9: (7=344KH%%  3 3&&tAAAvI/:6/B'DD __UE1q5 1 1FMM$qqq#v,/:3/?*AA$,$) + + + + RLL*WL===C   Jr-cHtj|\}}tj||j\}}t |}|jjdd|f}||j|z}||||jzdddl m }t||| }|j} t| it|}|d|d||fz|d|z|S) a Plot CCPR against one regressor. Generates a component and component-plus-residual (CCPR) plot. Parameters ---------- results : result instance A regression results instance. exog_idx : {int, str} Exogenous, explanatory variable. If string is given, it should be the variable name that you want to use, and you can use arbitrary translations as with a formula. ax : AxesSubplot, optional If given, it is used to plot in instead of a new figure being created. Returns ------- Figure If `ax` is None, the created figure. Otherwise the figure to which `ax` is connected. See Also -------- plot_ccpr_grid : Creates CCPR plot for multiple regressors in a plot grid. Notes ----- The CCPR plot provides a way to judge the effect of one regressor on the response variable by taking into account the effects of the other independent variables. The partial residuals plot is defined as Residuals + B_i*X_i versus X_i. The component adds the B_i*X_i versus X_i to show where the fitted line would lie. Care should be taken if X_i is highly correlated with any of the other independent variables. If this is the case, the variance evident in the plot will be an underestimate of the true variance. References ---------- http://www.itl.nist.gov/div898/software/dataplot/refman1/auxillar/ccpr.htm Examples -------- Using the state crime dataset plot the effect of the rate of single households ('single') on the murder rate while accounting for high school graduation rate ('hs_grad'), percentage of people in an urban area, and rate of poverty ('poverty'). >>> import statsmodels.api as sm >>> import matplotlib.pyplot as plt >>> import statsmodels.formula.api as smf >>> crime_data = sm.datasets.statecrime.load_pandas() >>> results = smf.ols('murder ~ hs_grad + urban + poverty + single', ... data=crime_data.data).fit() >>> sm.graphics.plot_ccpr(results, 'single') >>> plt.show() .. plot:: plots/graphics_regression_ccpr.py Nrhr) add_constantrvz*Component and component plus residual plotzResidual + %s*beta_%dz%s)r rLrMrNrrPrr6rostatsmodels.tools.toolsrr rrrrVrXrW) r*rZr8r]r^r`x1betarmodrs r+rr<s(|!"%%GC1(GMJJIx"7++G  AAAxK (B x( (FGGB&,,,444444 fll2&& ' ' + + - -C ZF v - - -CLL=>>>MM)Y,AABBBMM$"### Jr-ctj|}tj||j\}}||\}}nXt |dkr4t t jt |dz }d}nt |}d}d}t|D]x\}} |jj dd| f dkrd}4| |||dz|z } t|| | }| dy|dd ||d |S) a Generate CCPR plots against a set of regressors, plot in a grid. Generates a grid of component and component-plus-residual (CCPR) plots. Parameters ---------- results : result instance A results instance with exog and params. exog_idx : None or list of int The indices or column names of the exog used in the plot. grid : None or tuple of int (nrows, ncols) If grid is given, then it is used for the arrangement of the subplots. If grid is None, then ncol is one, if there are only 2 subplots, and the number of columns is two otherwise. fig : Figure, optional If given, this figure is simply returned. Otherwise a new figure is created. Returns ------- Figure If `ax` is None, the created figure. Otherwise the figure to which `ax` is connected. See Also -------- plot_ccpr : Creates CCPR plot for a single regressor. Notes ----- Partial residual plots are formed as:: Res + Betahat(i)*Xi versus Xi and CCPR adds:: Betahat(i)*Xi versus Xi References ---------- See http://www.itl.nist.gov/div898/software/dataplot/refman1/auxillar/ccpr.htm Examples -------- Using the state crime dataset separately plot the effect of the each variable on the on the outcome, murder rate while accounting for the effect of all other variables in the model. >>> import statsmodels.api as sm >>> import matplotlib.pyplot as plt >>> import statsmodels.formula.api as smf >>> fig = plt.figure(figsize=(8, 8)) >>> crime_data = sm.datasets.statecrime.load_pandas() >>> results = smf.ols('murder ~ hs_grad + urban + poverty + single', ... data=crime_data.data).fit() >>> sm.graphics.plot_ccpr_grid(results, fig=fig) >>> plt.show() .. plot:: plots/graphics_regression_ccpr_grid.py Nrgr'rr)rZr8rz&Component-Component Plus Residual Plotrkrlrrw)r ryrMrNrintrQceilrrPvarrzrrVrrr) r*rZrr]r^rr seen_constantrrr8s r+rrsf~  s # #C1(GMJJIx  uu x==1  H b 01122EEEMMEEMH%%3 = aaaf % ) ) + +q 0 0M  __UE1Q3}+< = =#"555 RLL9GLLLLC   Jr-c 8||}nd}tj|\}}|r]|j\|P|jjdddf|jjdddfg}n)td||}|dzz|dzzg} | |ddl m } Gfdd| } | || fi|} | | |j d| j| _|j d | j| _|r|||r|||S) a Plot a line given an intercept and slope. Parameters ---------- intercept : float The intercept of the line. slope : float The slope of the line. horiz : float or array_like Data for horizontal lines on the y-axis. vert : array_like Data for verterical lines on the x-axis. model_results : statsmodels results instance Any object that has a two-value `params` attribute. Assumed that it is (intercept, slope). ax : axes, optional Matplotlib axes instance. **kwargs Options passed to matplotlib.pyplot.plt. Returns ------- Figure The figure given by `ax.figure` or a new instance. Examples -------- >>> import numpy as np >>> import statsmodels.api as sm >>> np.random.seed(12345) >>> X = sm.add_constant(np.random.normal(0, 20, size=30)) >>> y = np.dot(X, [25, 3.5]) + np.random.normal(0, 30, size=30) >>> mod = sm.OLS(y,X).fit() >>> fig = sm.graphics.abline_plot(model_results=mod) >>> ax = fig.axes[0] >>> ax.scatter(X[:,1], y) >>> ax.margins(.1) >>> import matplotlib.pyplot as plt >>> plt.show() .. plot:: plots/graphics_regression_abline.py Nrz-specify slope and intercepty or model_resultsr)Line2Dc6eZdZfdZfdZfdZxZS)abline_plot..ABLine2DcVtj|i|d|_d|_dSN)super__init__id_xlim_callbackid_ylim_callback)selfargsr\ __class__s r+rz&abline_plot..ABLine2D.__init__5s5 EGG d -f - - -$(D !$(D ! ! !r-c|j}|jr|j|j|jr|j|jt dSr)axesr callbacks disconnectrrremove)rr8rs r+rz$abline_plot..ABLine2D.remove:slB$ ? ''(=>>>$ ? ''(=>>> GGNN     r-cP|d|}fd|D}|d}|}|dzz|dzzg}||||jjdS)NFcg|]}|u| Sr).0childrs r+ z@abline_plot..ABLine2D.update_datalim..EsDDDetmmummmr-rr)set_autoscale_on get_childrenget_xlimset_datar7canvasdraw) rr8childrenablinesablinerr_ interceptslopes ` r+update_datalimz,abline_plot..ABLine2D.update_datalimBs    & & &((HDDDD(DDDGQZF A1 )1Q4%<)+CDA OOAq ! ! ! I  ! ! # # # # #r-)__name__ __module__ __qualname__rrr __classcell__)rrrs@r+ABLine2Dr4sp ) ) ) ) )       $ $ $ $ $ $ $ $ $ $r-r xlim_changed ylim_changed)rr rLrrNrPminmaxrset_xlimmatplotlib.linesradd_linerconnectrrrhlinevline) rrhorizvert model_resultsr8r\rr]data_yrrlines `` r+rrs\ ~ KKMM !"%%GC (/ 5 9$)!!!Q$/3355$)!!!Q$/33557A%%*;LMM M 9 Ad5j"AaDJy$8 9FKKNNN('''''$$$$$$$$6$$$2 8Av ( ( ( (DKKL00ATUUDL00ATUUD     Jr-extra_params_doczresults: object Results for a fitted regression model. influence: instance The instance of Influence for model.皙?cooks0?c |} tj|\} }|dr| jd} nY|dr t j| jd} ntd|zt j | }|dzdz }| | z |z|z dz} || j }| d}|r| j } n| j } nt j| } d}dd lm}|jd |dz z |j}t j| |k}|t)|k}t j||}||| | | |jjj}|t5t7| }tjt j|d|t=|| t=| dz d z | dz d zd |}ddd}|j|fi||j di||j!di|| S)NcoordffzCriterion %s not understoodrg@zStudentized Residuals Residuals)stats?)srHrjrrn)rmrCLeverageInfluence Plot)r)r)"r rLlower startswithcooks_distancerQabsdffitsrptprhat_matrix_diagresid_studentized_externalresid_studentizedrscipyrtppfdf_residr, logical_orscatterrNrrrrrwhererrXrWrV)r* influenceexternalrH criterionr plot_alphar8leverageror\inflr]psize old_range new_rangeylabelrcutoff large_residlarge_leverage large_pointslabelsfonts r+_influence_plotr@Ysz D!"%%GC##E**D#A&    % %e , ,Dt{1~&&6BCCC u Ia$I UYY[[ I -i 7$ >E' }(  +3EE*EE 5!! W[[E!GW%5 6 6F&--&(Kw 7 77N=n==LJJx%zJ:::]  *F ~E ##  RXl33A6!(E22!E!Gb=.57R-@@) ! !B W - -DBM&!!D!!!BM%%%%%BL**T*** Jr-z@results : Results Results for a fitted regression model.c X|}t||f||||||d|} | S)N)r1rHr2rr3r8) get_influencer@) r*r1rHr2rr3r8r\r5ress r+rrsT  " "D '4 B(%$-D%/B B B:@ B BC Jr-zqresults: object Results for a fitted regression model influence: instance instance of Influence for modelc ddlm}m}tj|\}}|}|j} ||j} |j| dz| dfi||d| d| d| t|k} | d|dz z } tj| | k} |jjj}|!t%t'|j}tjtj| | d}tj||t1| dz| d gt'|jzd |d d }|dd|S)Nr)normzscorergrhzNormalized residuals**2rz)Leverage vs. Normalized residuals squaredrrrkrr)r8rrg333333?) scipy.statsrErFr rLr&ror6rWrXrVr,r+rQr#rNrrrrr)r/r-rrmargins)r*r0rHr8r\rErFr]r5r4ror<r:r;r>rss r+_plot_leverage_resid2rIs)(((((((!"%%GC D#H F4:  E BGE1Hh..v...MM+,,,MM*LL<===w 7 77N XXbqj ! !F&--&(K ]  *F ~GL))** HR]>;?? @ @ CE  UFD8,D,D$Xc',&7&77 "xH > > >BJJtT Jr-z:results : object Results for a fitted regression modelc L|}t||f||d|S)N)rHr8)rBrI)r*rHr8r\r5s r+rrs4  " "D $ Me M Mf M MMr-c|j}tj|\}}t|||||\}} || |dd|ddt |tr|} n |j|} | | d | |j d zd |S) N) resid_typeuse_glm_weights fit_kwargsrh333333?rHzAdded variable plotrkrlrz residuals) rNr rLr r6rVrrrrWrXrS) r* focus_exogrLrMrNr8rNr] endog_residfocus_exog_residxnames r+r#r#s ME!"%%GC'w 2<7F2<>>>"K! GG k3cG:::LL&L999*c""- ,MM%bM!!!MM%#l2M<<< Jr-c|j}tj||\}}t||}|jjdd|f}tj|\}}|||dd|ddt|tr|}n |j |}| |d| d d|S) NrhrOrPzPartial residuals plotrkrlrQrRComponent plus residual) rNr rMr!rPrLr6rVrrrrWrX) r*rSr8rN focus_colprfocus_exog_valsr]rVs r+r$r$s ME!3JFFJ  , ,Bm(I6O!"%%GCGGORCG000LL)GL<<<*c""- ,MM%bM!!!MM+"M555 Jr-zHresults : Results Results instance of a fitted regression model.Q?cv|j}tj||\}}t||||}|jdd|f}tj|\} }|||dd|dd||d | d d | S) N)r0 cond_meansrhrOrPzCERES residuals plotrkrlrQrRrX) rNr rMr"rPrLr6rVrWrX) r*rSr0r^r8rNrYpresidr[r]s r+r%r%s ME!3JFFJ ':D%/111FjI.O!"%%GCGGOVSG444LL''L:::MM*2M&&&MM+"M555 Jr-cH|j}t|tttfst d|jjztj ||\}}tt|j }t|}||t|}||jdd|f}||dz}t"j|d\} } } t#j| dk} | dd| f}|jdd|f} t#jt| |jdf}t|jdD]*}|dd|f}t/|| |d}||dd|f<+t#j|jdd|f|fd}|j}|}||j|fi|}|}|j|jz }t|ttfr'||jj|jz}|jd|kr1|t#j |dd|df|j |dz }|S) a Calculate the CERES residuals (Conditional Expectation Partial Residuals) for a fitted model. Parameters ---------- results : model results instance The fitted model for which the CERES residuals are calculated. focus_exog : int The column of results.model.exog used as the 'focus variable'. frac : float, optional Lowess smoothing parameter for estimating the conditional means. Not used if `cond_means` is provided. cond_means : array_like, optional If provided, the columns of this array are the conditional means E[exog | focus exog], where exog ranges over some or all of the columns of exog other than focus exog. If this is an empty nx0 array, the conditional means are treated as being zero. If None, the conditional means are estimated. Returns ------- An array containing the CERES residuals. Notes ----- If `cond_means` is not provided, it is obtained by smoothing each column of exog (except the focus column) against the focus column. Currently only supports GLM, GEE, and OLS models. z$ceres residuals not available for %sNrgư>rF)r0 return_sorted)axis)!rNrrrr rrrr rMrangerrrrrPmeanrQlinalgsvd flatnonzerorr}r concatenate_get_init_kwdsrOrrTfamilylinkderivdot)r*rSr0r^rNrYix_nfnnfpexogurvtiifcoljr;cfnew_exogklass init_kwargs new_model new_resultr_s r+r"r"*sD ME ec3_ - -3?1233 3"3JFFJ  #gn%% & &E KKE IIi e**C  111e8$ A9==**1b ^AH % %!!!R%z!!!Y,'Xs4yy%+a.9:: u{1~&& " "Aqqq!tBDt5AAAB!Jqqq!t  ~uz!!!U(3Z@qIIIH OE&&((Kek8;;{;;IJ[:2 2F%#s$$C%,#))**ABBB~a3"&!!!STT'*J,=cdd,CDDD Mr-c|j}|j|z }t|tt fr(||jj|j z}nBt|tttfrntdt|zt|tur|j|}n|}|j||jdd|fz}||zS)a: Returns partial residuals for a fitted model with respect to a 'focus predictor'. Parameters ---------- results : results instance A fitted regression model. focus col : int The column index of model.exog with respect to which the partial residuals are calculated. Returns ------- An array of partial residuals. References ---------- RD Cook and R Croos-Dabrera (1998). Partial residual plots in generalized linear models. Journal of the American Statistical Association, 93:442. z+Partial residuals for '%s' not implemented.N)rNrOpredictrrrrjrkrlrTr r r rtyperrrsrrP)r*rSrNrorY focus_vals r+r!r!s8 ME K'//++ +E%#s$$( "(()=>>> ECc? + +( F;;'(( ( J3$**:66  y)EJqqq)|,DDI u r-c|j}t|tttfst d|jjz|j}|j }tj ||\}}|dd|f} |!t|ttfrd}nd}t|j d} t| } | ||dd| f} |j| } |j} |}| || fi|}d| i}||||jd i|}t)|ddst d  t)||}n #t*$rt d |zwxYwd dlmcm}t|ttfre|rc|j|j}t9|d r ||jz}|| | |}n(|| | }|j}||fS)aO Residualize the endog variable and a 'focus' exog variable in a regression model with respect to the other exog variables. Parameters ---------- results : regression results instance A fitted model including the focus exog and all other predictors of interest. focus_exog : {int, str} The column of results.model.exog or a variable name that is to be residualized against the other predictors. resid_type : str The type of residuals to use for the dependent variable. If None, uses `resid_deviance` for GLM/GEE and `resid` otherwise. use_glm_weights : bool Only used if the model is a GLM or GEE. If True, the residuals for the focus predictor are computed using WLS, with the weights obtained from the IRLS calculations for fitting the GLM. If False, unweighted regression is used. fit_kwargs : dict, optional Keyword arguments to be passed to fit when refitting the model. Returns ------- endog_resid : array_like The residuals for the original exog focus_exog_resid : array_like The residuals for the focus predictor Notes ----- The 'focus variable' residuals are always obtained using linear regression. Currently only GLM, GEE, and OLS models are supported. z8model type %s not supported for added variable residualsNresid_devianceror start_params convergedTz>fit did not converge when calculating added variable residualsz '%s' residual type not availabler data_weightsr) rNrrrr rrrrPrOr rMrcr}rrrrirrgetattrAttributeError#statsmodels.regression.linear_model regression linear_modelrjweightsrTrrr ro)r*rSrLrMrNrNrPrOrYr[rs reduced_exogrrxr\rzrr{rTlmr lm_resultsrUs r+r r sR ME ec3_ - -3S1233 3 :D KE!3JFFJ 111i<(O ec3Z ( ( !)JJ J tz!}  B bBFF92;L>"%L OE  ! ! # #Fe\44V44I L )D J&&&&J :{D 1 1[YZZZJj*55 JJJ;jHIIIJ544444444%#s$$AA,&&w';<< 5. ) ) 3 22GVVO\7CCGGII VVO\::>>@@ !' ( ((s E11F)rr.)NNTr)NNN)NNNNNNr)TrrrrNNN)TrrrrN)rN)NTNN)r\NN)r\N)NTN)8__doc__statsmodels.compat.pandasrstatsmodels.compat.pythonrrnumpyrQrrpatsyr3statsmodels.genmod.generalized_estimating_equationsr+statsmodels.genmod.generalized_linear_modelrstatsmodels.graphicsr *statsmodels.nonparametric.smoothers_lowessr rr r r &statsmodels.sandbox.regression.predstdrrr_regressionplots_docrrrrr__all__r,rrrrrrrrrformatr@rrIrr#r$r%r"r!r rr-r+rs  /.....22222222CCCCCC;;;;;;&&&&&&================EEEEEE888888 = = =444 8]]]]@bbbbJ!!!:7;"$BddddNccccLPPPPf\\\\~>B'+hhhhV  $  $ I IG(H I IJJ >ACG)-;;; JJ ;|  $  $ K KI(J K KLLAH/3LL  * # *>,? @ @AA  AA :  * # *D,E F FGGNNNGGN  "D&E EFF9=BFFF6  %D)E EFFFF2  #!'" "##EI  ##.WWWWr...`;?;?])])])])])])r-