M/Ph?lddlZddlmZddlmZddlmZddl m Z m Z ddl m Z GddZGd d Zdd ZGd d ZdZdZ ddZGddZddZ d dZ d!dZdZd"dZ d"dZdS)#N)f)t)stats)clean0fullrank) multipletestscBeZdZdZ d dZd dZdZdZd dZd d Z dS)ContrastResultsaX Class for results of tests of linear restrictions on coefficients in a model. This class functions mainly as a container for `t_test`, `f_test` and `wald_test` for the parameters of a model. The attributes depend on the statistical test and are either based on the normal, the t, the F or the chisquare distribution. N皙?c ||_|Zd|_||_|j|_||_||_t |_||f|_t j ||||_ n|nd|_||_ ||_||_ ||_t|_|f|_|j tj||dz|_ n0d|vr|d|_|d|_|dx|_ } ||_ t!t"|j|_|dd|_|jdkr-|j |j||_ ||_n tj| tj|_ tj| } |j tj| | dz|j | <ntj|_ tj|j |_ |j2d t/t1|jD|_dSd|_dS) NFr statistic distribution dist_argschi2cg|]}d|zS)zc%dr).0iis Z/var/www/html/test/jupyter/venv/lib/python3.11/site-packages/statsmodels/stats/contrast.py z,ContrastResults.__init__..DsIII2EBJIII)effectrfvaluerdf_denomdf_numfdistdistrsfpvaluetvaluesd student_tnpabsgetattrrget full_likenanisnansqueezerangelenc_names) selfrr r#rrralphakwdsvaluenot_nans r__init__zContrastResults.__init__s = #D DK![DN$DM DKDI$h/DN(1fh77DKK ] #D DKDNDG$DM!DI&[DN),,rvayy(;;a?DKK D $^ 4D !+.DN"&{"3 3DK%DGt'899DI!XXk266DN F**"ill4>8DD (  l5"&99 8E??*'+y||BF5>4J4J'K'Ka'O G$$&DKj-- ; "IIs4;7G7G1H1HIIIDLLLDLLLrc|jX|jjd|dz z g|jR}|j||jzz }|j||jzz}t j||fStd)a& Returns the confidence interval of the value, `effect` of the constraint. This is currently only available for t and z tests. Parameters ---------- alpha : float, optional The significance level for the confidence interval. ie., The default `alpha` = .05 returns a 95% confidence interval. Returns ------- ci : ndarray, (k_constraints, 2) The array has the lower and the upper limit of the confidence interval in the columns. Ng@z!Confidence Interval not available)rrppfrr#r% column_stackNotImplementedError)r0r1qloweruppers rconf_intzContrastResults.conf_intHsx$ ; " a%"*n>t~>>>AK!dg+-EK!dg+-E?E5>22 2%&IJJ JrcN|SN)summary__str__r0s rrBzContrastResults.__str__cs||~~%%'''rcZt|jdz|zSN str __class__rBrCs r__repr__zContrastResults.__repr__f$4>""T)DLLNN::rc |j}|d}n|dkrd}|jdk}d}||j}ddlm}t j|j}|||j|j|j || |f|||||}|St|d r*d t|j |j|j|jfzS|jd krd |j|j |j|jfzSd |j d|jdS)aSummarize the Results of the hypothesis test Parameters ---------- xname : list[str], optional Default is `c_##` for ## in the number of regressors alpha : float significance level for the confidence intervals. Default is alpha = 0.05 which implies a confidence level of 95%. title : str, optional Title for the params table. If not None, then this replaces the default title Returns ------- smry : str or Summary instance This contains a parameter results table in the case of t or z test in the same form as the parameter results table in the model results summary. For F or Wald test, the return is a string. NzTest for Constraintsr constraintsr)summary_params)ynamexnameuse_ttitler1rz0rz9z)rrr/statsmodels.iolib.summaryrOr% atleast_1dr!r#rr>hasattrreprrrr) r0rQr1rSrRrPrOpvaluessumms rrAzContrastResults.summaryisN, ; "}."&#-EE}  @ @ @ @ @ @mDK00G!>4dgt~#*DMM%,@,@#B(-U%(-U<<r:)r0rQr1rRrPr\rZs r summary_framezContrastResults.summary_frames ; "&#-EE}  F F F F F F''t{DG)- )-u)=)=)?FK/4E/4 666D K&&BCC Cr)NNNNNNr )r )Nr N)Nr ) __name__ __module__ __qualname____doc__r5r>rBrJrAr]rrrr r sGK$(1 1 1 1 fKKKK6(((;;;72727272tDDDDDDrr c:eZdZdZdZeeZdZdZdS)ContrastaZ This class is used to construct contrast matrices in regression models. They are specified by a (term, design) pair. The term, T, is a linear combination of columns of the design matrix. The matrix attribute of Contrast is a contrast matrix C so that colspan(dot(D, C)) = colspan(dot(D, dot(pinv(D), T))) where pinv(D) is the generalized inverse of D. Further, the matrix Tnew = dot(C, D) is full rank. The rank attribute is the rank of dot(D, dot(pinv(D), T)) In a regression model, the contrast tests that E(dot(Tnew, Y)) = 0 for each column of Tnew. Parameters ---------- term : array_like design : array_like Attributes ---------- contrast_matrix Examples -------- >>> import statsmodels.api as sm >>> from statsmodels.stats.contrast import Contrast >>> import numpy as np >>> np.random.seed(54321) >>> X = np.random.standard_normal((40,10)) # Get a contrast >>> new_term = np.column_stack((X[:,0], X[:,2])) >>> c = Contrast(new_term, X) >>> test = [[1] + [0]*9, [0]*2 + [1] + [0]*7] >>> np.allclose(c.contrast_matrix, test) True Get another contrast >>> P = np.dot(X, np.linalg.pinv(X)) >>> resid = np.identity(40) - P >>> noise = np.dot(resid,np.random.standard_normal((40,5))) >>> new_term2 = np.column_stack((noise,X[:,2])) >>> c2 = Contrast(new_term2, X) >>> print(c2.contrast_matrix) [ -1.26424750e-16 8.59467391e-17 1.56384718e-01 -2.60875560e-17 -7.77260726e-17 -8.41929574e-18 -7.36359622e-17 -1.39760860e-16 1.82976904e-16 -3.75277947e-18] Get another contrast >>> zero = np.zeros((40,)) >>> new_term3 = np.column_stack((zero,X[:,2])) >>> c3 = Contrast(new_term3, X) >>> test2 = [0]*2 + [1] + [0]*7 >>> np.allclose(c3.contrast_matrix, test2) True cXt|ds||jS)z3 Gets the contrast_matrix property _contrast_matrix)rWcompute_matrixrerCs r _get_matrixzContrast._get_matrixs2t/00 "    ! ! !$$rcjtj||_tj||_dSr@)r%asarraytermdesign)r0rjrks rr5zContrast.__init__s(Jt$$ j(( rc|j}|jdkr |dddf}t||_|j|_t |j|j|_ |jj d|_ dS#d|_ YdSxYw)z Construct a contrast matrix C so that colspan(dot(D, C)) = colspan(dot(D, dot(pinv(D), T))) where pinv(D) is the generalized inverse of D=design. r7N) rjndimrTrkDcontrastfromcolsrematrixshaperank)r0rns rrfzContrast.compute_matrix s I 6Q;;!!!D& A 0 @ @  )!,DIII DIIIIs A88 BN) r^r_r`rargpropertycontrast_matrixr5rfrrrrcrcs]??@%%%h{++O)))rrcctj|}tj|}|j\}}|jd|kr |jd|krtd|tj|}|jd|krtj||j}n4|}tj|tj||jj}tj||j}t|jdkr |df|_tj ||jdkr)t|}tj||j}tj |S)a From an n x p design matrix D and a matrix L, tries to determine a p x q contrast matrix C which determines a contrast of full rank, i.e. the n x q matrix dot(transpose(C), pinv(D)) is full rank. L must satisfy either L.shape[0] == n or L.shape[1] == p. If L.shape[0] == n, then L is thought of as representing columns in the column space of D. If L.shape[1] == p, then L is thought of as what is known as a contrast matrix. In this case, this function returns an estimable contrast corresponding to the dot(D, L.T) Note that this always produces a meaningful contrast, not always with the intended properties because q is always non-zero unless L is identically 0. That is, it produces a contrast that spans the column space of L (after projection onto the column space of D). Parameters ---------- L : array_like D : array_like rr7zshape of L and D mismatched) r%rirr ValueErrorlinalgpinvdotrnr. matrix_rankrr,)LropseudonpCLps rrprp s4< 1 A 1 A 7DAqwqzQ171:??6777 ~""wqzQ F61     F626!QS>> * * , 13B 28}}q6 yR  BHQK// b\\ F62    :a==rcBeZdZ ddZedZdZdZdZdS)WaldTestResultsNc8||_||_||_||_|W|dj|_|dj|_|dj|_|jdkr|dj|_dSdS|jdkr$tj |_ |jd|_n@|jdkr&tj |_ |j\|_|_ntd|,|j j tj|g|R|_dS||_dS) Nrr! df_constraintr rrrz)only F and chi2 are possible distribution)tablerrrvaluesrYdf_constraintsrrrrrrwr r%r&)r0rrrrrYs rr5zWaldTestResults.__init___s (""  ";/6DN ?1DL"'"8"?D  C'' %j 1 8 (' F**!J &*nQ&7##"c))!G 59^2#T]]!!LMMM+ty|BF9,=,=J JJJ & rcnd|jz}|j|dg}|jdkr|d|S)z'column names for summary table zP>%sz df constraintr zdf denom)rappend)r0pr_test col_namess rrzWaldTestResults.col_namessH 4,,&A   # #   Z ( ( (rct|dr|jStt|jj|j}|j||_|jS)N_dframe)columns) rWrdictziprrrrenamedframe)r0renamings rr]zWaldTestResults.summary_frames\ 4 # # < DJ.??@@j'''99 {rcN|Sr@)r] to_stringrCs rrBzWaldTestResults.__str__s !!##--///rcZt|jdz|zSrErGrCs rrJzWaldTestResults.__repr__rKr)NN) r^r_r`r5rtrr]rBrJrrrrr\s{BF ' ' ' 'DX000;;;;;rrc\tj|d}fdt|D}|S)z8helper function for labels for pairwise comparisons r7cNg|]!}|dd|d"S)r7-rr)rname level_namess rrz%_get_pairs_labels..sI///T!W%>> DG(<>>///r)r% triu_indicesr)k_levelr idx_pairs_alllabelss ` r_get_pairs_labelsrsHOGQ//M////}-///F Mrc|dz }tj|d}t|d}tj||f}d|tj||df<d|tj||df<tj|}tj||fd}|jd} tj| |f} || dd|||zf<| S)acreate pairwise contrast for reference coding currently not used, using encoding contrast matrix is more general, but requires requires factor information from patsy design_info. Parameters ---------- k_params : int number of parameters k_level : int number of levels or categories (including reference case) idx_start : int Index of the first parameter of this factor. The restrictions on the factor are inserted as a block in the full restriction matrix starting at column with index `idx_start`. Returns ------- contrasts : ndarray restriction matrix with k_params columns and number of rows equal to the number of restrictions. r7r)axisN)r%rr.zerosarangeeye concatenaterr) k_paramsr idx_start k_level_m1 idx_pairskc_pairs c_referenceck_all contrastss r_contrast_pairsrs21J A..I IaLAh:''G*,GBIaLL)A, &'*+GBIaLL)A, &'&$$K  W-A666A GAJE%*++I78IaaaY3334 rhsr c||}||}t|tur|g}|D]5}t |j||} | d|d|z<| d|d|z<6|S)aperform t_test and add multiplicity correction to results dataframe Parameters ---------- result results instance results of an estimated model contrasts : ndarray restriction matrix for t_test method : str or list of strings method for multiple testing p-value correction, default is'hs'. alpha : float significance level for multiple testing reject decision. ci_method : None not used yet, will be for multiplicity corrected confidence intervals contrast_names : {list[str], None} If contrast_names are provided, then they are used in the index of the returned dataframe, otherwise some generic default names are created. Returns ------- res_df : pandas DataFrame The dataframe contains the results of the t_test and additional columns for multiplicity corrected p-values and boolean indicator for whether the Null hypothesis is rejected. )rQ)methodr1r7z pvalue-%srz reject-%s)t_testr]typelistrr!) resultrrr1 ci_methodcontrast_namesttres_dfmethmts r t_test_multirs6 y ! !B   N  3 3F F||4++ 29T ? ? ?%'U{T!"%'U{T!"" MrceZdZdZdZdS)MultiCompResultzdclass to hold return of t_test_pairwise currently just a minimal class to hold attributes. c :|j|dSr@)__dict__update)r0kwargss rr5zMultiCompResult.__init__s V$$$$$rN)r^r_r`rar5rrrrrs-%%%%%rrcz|j\}}tj||f}|||dd|||zf<n ||dd|f<|S)aqhelper function to expand constraints to a full restriction matrix Parameters ---------- contrasts : ndarray restriction matrix for t_test k_params : int number of parameters idx_start : int Index of the first parameter of this factor. The restrictions on the factor are inserted as a block in the full restriction matrix starting at column with index `idx_start`. index : slice or ndarray Column index if constraints do not form a block in the full restriction matrix, i.e. if parameters that are subject to restrictions are not consecutive in the list of parameters. If index is not None, then idx_start is ignored. Returns ------- contrasts : ndarray restriction matrix with k_params columns and number of rows equal to the number of restrictions. N)rrr%r)rrrindexk_ck_prs r_embed_constraintsrs\4HC #x!!A },5!!!YS ( ())!!!U( Hrpairwisec|}|j\}}ddlmcmcm}|dvr|| }nt d||} |"|tdt| ||} | S)a helper function to create constraints based on encoding matrix Parameters ---------- encoding_matrix : ndarray contrast matrix for the encoding of a factor as defined by patsy. The number of rows should be equal to the number of levels or categories of the factor, the number of columns should be equal to the number of parameters for this factor. comparison : str Currently only 'pairwise' is implemented. The restriction matrix can be used for testing the hypothesis that all pairwise differences are zero. k_params : int number of parameters idx_start : int Index of the first parameter of this factor. The restrictions on the factor are inserted as a block in the full restriction matrix starting at column with index `idx_start`. Returns ------- contrast : ndarray Contrast or restriction matrix that can be used in hypothesis test of model results. The number of columns is k_params. rN)rpwpairsz!currentlyonly pairwise comparisonz3if k_params is not None, then idx_start is required) rr#statsmodels.sandbox.stats.multicompsandboxr multicompcontrast_allpairsr:rzrwr) encoding_matrix comparisonrrcmrrmcc_allrs r_constraints_factorr&s: B8LGS444444444444000%%g...!"EFFF " I  ()) )&y(IFF rFc@|jjj}|j|}|j|}|j|j} |s't|j dkrtd|j d} |j | j } |Bt|t| kr|} ntdt| zt| } |j |dj| j} t|j}t#| | }ddlmcmcm}||  }|| }t1||| }t3|||d||}t5||||| }|S)a Perform pairwise t_test with multiple testing corrected p-values. This uses the formula design_info encoding contrast matrix and should work for all encodings of a main effect. Parameters ---------- result : result instance The results of an estimated model with a categorical main effect. term_name : str name of the term for which pairwise comparisons are computed. Term names for categorical effects are created by patsy and correspond to the main part of the exog names. method : {str, list[str]} multiple testing p-value correction, default is 'hs', see stats.multipletesting alpha : float significance level for multiple testing reject decision. factor_labels : {list[str], None} Labels for the factor levels used for pairwise labels. If not provided, then the labels from the formula design_info are used. ignore : bool Turn off some of the exceptions raised by input checks. Returns ------- MultiCompResult The results are stored as attributes, the main attributes are the following two. Other attributes are added for debugging purposes or as background information. - result_frame : pandas DataFrame with t_test results and multiple testing corrected p-values. - contrasts : matrix of constraints of the null hypothesis in the t_test. Notes ----- Status: experimental. Currently only checked for treatment coding with and without specified reference level. Currently there are no multiple testing corrected confidence intervals available. r7z%interaction effects not yet supportedrNz0factor_labels has the wrong length, should be %d)rrr1r) result_framerrjcontrast_labelsterm_encoding_matrix)modeldata design_info term_namesrterms term_slicesstartr.factorsrw factor_infos categories term_codingscontrast_matricesrqparamsrrrrrrrzrrr)r term_namerr1 factor_labelsignoredesinfoterm_idxrjrfactorcatrrrrr c_all_pairs contrasts_subrrress rt_test_pairwiserUsbl+G!'' 22H = "D#D)/I Bc$,''!++@AAA \!_F  v & 1C }  S ) )CCORUVYRZRZZ[[ [#hhG  d #A & 8 @ GB6=!!H w , ,F444444444444''000KOOB''M"=(IFFI &)Fd %f>>>F v$-#*0/1  3 3 3C Jrc ||z}||z}||z S)zoffset to the value of a linear constraint for new params usage: (cm, v) is original constraint vo = offset_constraint(cm, res2.params, params_alt) fs = res2.wald_test((cm, v + vo)) nc = fs.statistic * fs.df_num r)r_matrix params_est params_altdiff_estdiff_alts r_offset_constraintrs#*$H*$H h rTc|||z|z }||}|r$|tj||z}n)|tjtj|z }|S)a!Moncentrality parameter for a wald test in model results The null hypothesis is ``diff = r_matrix @ params - value = 0`` Parameters ---------- params : ndarray parameters of the model at which to evaluate noncentrality. This can be estimated parameters or parameters under an alternative. r_matrix : ndarray Restriction matrix or contrasts for the Null hypothesis value : None or ndarray Value of the linear combination of parameters under the null hypothesis. If value is None, then it will be replaced by zero. results : Results instance of a model The results instance is used to compute the covariance matrix of the linear constraints using `cov_params. diff : None or ndarray If diff is not None, then it will be used instead of ``diff = r_matrix @ params - value`` joint : bool If joint is True, then the noncentrality parameter for the joint hypothesis will be returned. If joint is True, then an array of noncentrality parameters will be returned, where elements correspond to rows of the restriction matrix. This correspond to the `t_test` in models and is not a quadratic form. Returns ------- nc : float or ndarray Noncentrality parameter for Wald tests, correspondig to `wald_test` or `t_test` depending on whether `joint` is true or not. It needs to be divided by nobs to obtain effect size. Notes ----- Status : experimental, API will likely change N)r) cov_paramsr%rxsolvesqrtdiag)rrr3resultsdiffjointcov_cncs rwald_test_noncentrstR |& 5(     1 1E , BIOOE400 0 BGBGENN++ + Irc|d}|||z|z }|}|||j}|r$|tj||z}n)|tjtj|z }|S)anoncentrality parameter for a wald test The null hypothesis is ``diff = r_matrix @ params - value = 0`` Parameters ---------- params : ndarray parameters of the model at which to evaluate noncentrality. This can be estimated parameters or parameters under an alternative. r_matrix : ndarray Restriction matrix or contrasts for the Null hypothesis value : None or ndarray Value of the linear combination of parameters under the null hypothesis. If value is None, then it will be replace by zero. cov_params : ndarray covariance matrix of the parameter estimates diff : None or ndarray If diff is not None, then it will be used instead of ``diff = r_matrix @ params - value`` joint : bool If joint is True, then the noncentrality parameter for the joint hypothesis will be returned. If joint is True, then an array of noncentrality parameters will be returned, where elements correspond to rows of the restriction matrix. This correspond to the `t_test` in models and is not a quadratic form. Returns ------- nc : float or ndarray Noncentrality parameter for Wald tests, correspondig to `wald_test` or `t_test` depending on whether `joint` is true or not. It needs to be divided by nobs to obtain effect size. Notes ----- Status : experimental, API will likely change Nr)rzrnr%rxrrr) rrr3rrrrrrs rwald_test_noncent_genericr sR } |& 5(A EE*   ! !!# & &E , BIOOE400 0 BGBGENN++ + Irr@)rr NN)rNN)rr NF)NT)numpyr% scipy.statsrrrr$scipyrstatsmodels.tools.toolsrrstatsmodels.stats.multitestrr rcrprrrrrrrrrrr rrrrs """"""&&&&&&44444444555555mDmDmDmDmDmDmDmDbbbbbbbbbJ8888x?;?;?;?;?;?;?;?;J'''THL $$$$$N%%%%%%%%     FJN"&,,,,^;?/4QQQQh    1111hIM$(555555r