!gYddlmZddlZddlZddlmZmZddlZddl m Z ddl m Z m Z mZddlmZddlmZdZd Zd Zd Zd Zd ZdZdZGddZGddZedk(r\ddlZddlmZej@ddgdZ!ede!jEZ#ede!jEZ$ee#dZ%yy))lrangeN) DataFrameIndex)stats)_has_intercept_intercept_idx_remove_intercept_patsy)summary2)OLSc||jS|dk(r |jS|dk(r |jS|dk(r |jS|dk(r |jSt d|z)Nhc0hc1hc2hc3z robust options %s not understood) cov_paramscov_HC0cov_HC1cov_HC2cov_HC3 ValueError)modelrobusts T/var/www/dash_apps/app1/venv/lib/python3.12/site-packages/statsmodels/stats/anova.py_get_covariancersk ~!! 5}} 5}} 5}} 5}};fDEEc 4|jdd}|jdd}|jdd}|jdd}|r|j}|jj}|jj}|j d}|jj } |jjj} |jj} t| jt| z dz} d |z} d d d || g}ttj| d f|}|dvrt!||||| || || | S|dvrt#|| | || |S|dvrt%|| | || |S|dvr t'dt)dt+|z)a9 Anova table for one fitted linear model. Parameters ---------- model : fitted linear model results instance A fitted linear model typ : int or str {1,2,3} or {"I","II","III"} Type of sum of squares to use. **kwargs** scale : float Estimate of variance, If None, will be estimated from the largest model. Default is None. test : str {"F", "Chisq", "Cp"} or None Test statistics to provide. Default is "F". Notes ----- Use of this function is discouraged. Use anova_lm instead. testFscaleNtyprrzPR(>%s)dfsum_sqmean_sqcolumnsr!I)II)III)IVzType IV not yet implementedzType %s not understood)getlowerrendogexogshape endog_namesdata design_info exog_nameslentermsrrnpzerosanova1_lm_singleanova2_lm_singleanova3_lm_singleNotImplementedErrorrstr)rkwargsrrr rr2r3nobs response_namer7r8n_rowspr_testnamestables r anova_singlerI#s. ::fc "D JJw %E **UA C ZZ$ 'F  KK  E ;;  D ::a=DKK++M++""..K''J+##$~k'BBQFF$G 8Yg 6E bhh{+U ;E hudD+u &gv? ? {FD' &( ( {FD' &( ( !"?@@1CH<==rc t|dd} | Btjj|\} } tj| j |} tj t|jt|jf} |jDcgc]}|j|}}t|D] \}}d| ||f<tj| | dz}t|}||}tj|j}||}|j}t!|dgz|_tj$| |j'd|f|j(|ddgf<|j*|j,f|j(dddgf<|dk(r|d|dz |j*|j,z z ||<t.j0j3|d|d|j,||<tj4tj4f|j(d||gf<|d|dz |d <|Scc}w) a Anova table for one fitted linear model. Parameters ---------- model : fitted linear model results instance A fitted linear model **kwargs** scale : float Estimate of variance, If None, will be estimated from the largest model. Default is None. test : str {"F", "Chisq", "Cp"} or None Test statistics to provide. Default is "F". Notes ----- Use of this function is discouraged. Use anova_lm instead. effectsNr!r*Residualr"r#rr$)getattrr;linalgqrdotTr<r9r: column_names term_namesslice enumeraterarraytolistrindexc_sumlocssrdf_residrfsfnan)rr2r3rCr7rHrErrFrrKqrarrnameslicesislice_r#idxrSrXs rr=r=_s.eY-Giill4 !&&e$ ((C ))*C 0H0H,IJ KC2=2H2H I$k% IF If%&AvIVVC! $F  %C SD\F+001JSD!J    E ,-EK)+sC4y}}Q/?/G)HEIIedH%%&-2YY-FEIIj8D/)* s{h%+5 ENN24d E#Jd $)NN4g13 *tWo-.Xt4E) L/JsIc6|jdd}t|}dd||g}ttj|df|}t |d} t ||} g} g} t |D]\} }|j|}t|j|j}g}t|j}|D]}t|j}|j|s*||k(r0|j|}|jt|j|j|jt|j|jtj|j j"j$d|}tj|j j"j$d|}|j&rtj(tj(|| |j*}ddlm}|j1|\}}|j$d|j$dz }tj(|dd| dfj*|}n|}|j$d}|d k(re|j3|| }|j4x|j6|j8| |f<}|j:|j6|j8| |f<||j6|j8| df<| j=|j| j=|j?tA| d gz|_|jBtjD| |j j"j$ddzgz}|||dz|jFz|jHz }||d<|jF|jHtjJtjJf|j6d dd||gf<|S) a Anova type II table for one fitted linear model. Parameters ---------- model : fitted linear model results instance A fitted linear model **kwargs** scale : float Estimate of variance, If None, will be estimated from the largest model. Default is None. test : str {"F", "Chisq", "Cp"} or None Test statistics to provide. Default is "F". Notes ----- Use of this function is discouraged. Use anova_lm instead. Type II Sum of Squares compares marginal contribution of terms. Thus, it is not particularly useful for models with significant interaction terms. Nr#r"r.r&r!r)rNrcov_prL)&r:r rr;r<rrUrTrstartstopsetfactorsissubsetextendeyerr3r4sizerPrQscipyrNrOf_testfvaluer[rXpvalueappendrdrilocargsortr\r]r`)rr7rErrFr terms_inforGrHcov robust_cov col_orderrXrftermcolsL1L2term_sett other_setcolLVLrN orth_compl_rbL12r^ test_valuer\s rr>r>sg2""1%J(4J tT7 +E bhh{+u =E % &C /JI EZ('"4  & DJJ * t||$ 7AAIII  +H 4I!''* &CHH56 &CHH56  7VVEKK$$**1- .r 2 VVEKK$$**1- .r 2 77&&:.rtt4C $!99S>LJq bhhqk)A&&AqbcE*,,b1CC A 3; S 3A;<88 CEIIekk!nd* +j12EIIekk!ng- .+, %++a.$&'$ TYY[!O'"R ,-EK JJrzz)u{{/?/?/E/Ea/H/J.K"KL ME +d #eii / >CE(O=BYYFFBFF=LEIIj8D$889 Lrc|t|z }|j}dd||g}ttj|df|}t ||} g} g} t |D]\} } |j| }tj|jjjd|}|}|jd}|dk(re|j|| }|jx|j|j| |f<}|j |j|j| |f<||j|j| df<| j#| j% t'| d gz|_|||dz|j(z|j*z }||d<|j(|j*tj,tj,f|jd dd||gf<|S) Nr#r"r.r&r!rrrjrL)rr:rr;r<rrUrTrrrr3r4rurvr[rXrwrxrdrr\r]r`)rr7rErrFrr{rGrHr|r~rXrfrrrrrbr^rr\s rr?r?s n[))F""J tT7 +E bhh{+u =E % (CI EZ("4  & VVEKK$$**1- .t 4 HHQK 3; S ,A;<88 CEIIekk!nd* +j12EIIekk!ng- .+, %++a.$&' TYY[!"" ,-EK +d #eii / >CE(O=BYYFFBFF=LEIIj8D$889 Lrc|jdd}t|dk(r|d}t|fi|S|dvrtdt |z|jdd}|jdd }t|}d |z}d d d d||g}t t j|df|} |s|dj}|D cgc]} | jc} | d <|D cgc]} | jc} | d <t j| d j | j| jdd d f<| d j | d<|dk(rn| d| d z |z | d<tj j#| d| d | d | |<t j$| j| dj'|f<| Scc} wcc} w)a Anova table for one or more fitted linear models. Parameters ---------- args : fitted linear model results instance One or more fitted linear models scale : float Estimate of variance, If None, will be estimated from the largest model. Default is None. test : str {"F", "Chisq", "Cp"} or None Test statistics to provide. Default is "F". typ : str or int {"I","II","III"} or {1,2,3} The type of Anova test to perform. See notes. robust : {None, "hc0", "hc1", "hc2", "hc3"} Use heteroscedasticity-corrected coefficient covariance matrix. If robust covariance is desired, it is recommended to use `hc3`. Returns ------- anova : DataFrame When args is a single model, return is DataFrame with columns: sum_sq : float64 Sum of squares for model terms. df : float64 Degrees of freedom for model terms. F : float64 F statistic value for significance of adding model terms. PR(>F) : float64 P-value for significance of adding model terms. When args is multiple models, return is DataFrame with columns: df_resid : float64 Degrees of freedom of residuals in models. ssr : float64 Sum of squares of residuals in models. df_diff : float64 Degrees of freedom difference from previous model in args ss_dff : float64 Difference in ssr from previous model in args F : float64 F statistic comparing to previous model in args PR(>F): float64 P-value for significance comparing to previous model in args Notes ----- Model statistics are given in the order of args. Models must have been fit using the formula api. See Also -------- model_results.compare_f_test, model_results.compare_lm_test Examples -------- >>> import statsmodels.api as sm >>> from statsmodels.formula.api import ols >>> moore = sm.datasets.get_rdataset("Moore", "carData", cache=True) # load >>> data = moore.data >>> data = data.rename(columns={"partner.status" : ... "partner_status"}) # make name pythonic >>> moore_lm = ols('conformity ~ C(fcategory, Sum)*C(partner_status, Sum)', ... data=data).fit() >>> table = sm.stats.anova_lm(moore_lm, typ=2) # Type 2 Anova DataFrame >>> print(table) r r!rr(z6Multiple models only supported for type I. Got type %srrrNzPr(>%s)r]r\df_diffss_diffr&)r0r9rIrrArr;r<rr\r]diffvaluesr[rXrr^r_r`isnull) argsrBr rrrn_modelsrFrGrHmdls ranova_lmrsL **UA C 4yA~QE,V,, ('),S23 3 ::fc "D JJw %E4yH$G  9dG DE bhh!}-u =E R'+,CGG,E%L156#6E*-/WWU:5F5M5M-N,NEIIekk!"oy()e ))++E) s{9%i(885@c E#Ji0@$)*$57g35&& %*##%w./ L-6s GGcZtjdg|z}|D] }||}d||<|S)NTF)r;rV)reslices_to_excludenindrss r _not_slicers> ((D6!8 C! 4LA Jrct|||jd}||}tj||dd|fj |}|j j |}t |t |z }||fS)ah Residual sum of squares of OLS model excluding factors in `keys` Assumes x matrix is orthogonal Parameters ---------- y : array_like dependent variable x : array_like independent variables term_slices : a dict of slices term_slices[key] is a boolean array specifies the parameters associated with the factor `key` params : ndarray OLS solution of y = x * params keys : keys for term_slices factors to be excluded Returns ------- rss : float residual sum of squares df : int degrees of freedom r!N)rr4r;subtractrPrQr9) yx term_slicesparamskeysrparams1r\r]s r_ssr_reduced_modelrsq4 [$ 3CSkG ++a1c6w/ 0C %%))C.C1vG $H =rc.eZdZdZ ddZdZdZdZy)AnovaRMa Repeated measures Anova using least squares regression The full model regression residual sum of squares is used to compare with the reduced model for calculating the within-subject effect sum of squares [1]. Currently, only fully balanced within-subject designs are supported. Calculation of between-subject effects and corrections for violation of sphericity are not yet implemented. Parameters ---------- data : DataFrame depvar : str The dependent variable in `data` subject : str Specify the subject id within : list[str] The within-subject factors between : list[str] The between-subject factors, this is not yet implemented aggregate_func : {None, 'mean', callable} If the data set contains more than a single observation per subject and cell of the specified model, this function will be used to aggregate the data before running the Anova. `None` (the default) will not perform any aggregation; 'mean' is s shortcut to `numpy.mean`. An exception will be raised if aggregation is required, but no aggregation function was specified. Returns ------- results : AnovaResults instance Raises ------ ValueError If the data need to be aggregated, but `aggregate_func` was not specified. Notes ----- This implementation currently only supports fully balanced designs. If the data contain more than one observation per subject and cell of the design, these observations need to be aggregated into a single observation before the Anova is calculated, either manually or by passing an aggregation function via the `aggregate_func` keyword argument. Note that if the input data set was not balanced before performing the aggregation, the implied heteroscedasticity of the data is ignored. References ---------- .. [*] Rutherford, Andrew. Anova and ANCOVA: a GLM approach. John Wiley & Sons, 2011. Nc||_||_||_d|vr td||_| t d||_|dk(r tjj|_ n||_ |j|j|g|zs*|j|jn d}t||jy)NCzSFactor name cannot be 'C'! This is in conflict with patsy's contrast function name.z)Between subject effect not yet supported!mean)subsetzThe data set contains more than one observation per subject and cell. Either aggregate the data manually, or pass the `aggregate_func` parameter.)r6depvarwithinrbetweenr@subjectpdSeriesraggregate_funcequalsdrop_duplicates _aggregate_check_data_balanced)selfr6rrrrrmsgs r__init__zAnovaRM.__init__s   &=DE E  %'78 8 V #"$))..D "0D {{4//y67I/JK"".!A!o% !!#rc|jj|jg|jzd|jj |j |_y)NF)as_index)r6groupbyrrraggrrs rrzAnovaRM._aggregatesQYYgt||nt{{:',..2kk;c$--. rcRd}|jD]+}|t|j|jz}-i}t |jj dD]`}g}|jD]-}|j |j|j|/t|}||vr ||dz||<\d||<bd}t||k7r t||}|D]}|||k7s t||jj d||zkDr tdy)zraise if data is not balanced This raises a ValueError if the data is not balanced, and returns None if it is balance Return might change r!rzData is unbalanced.z9There are more than 1 element in a cell! Missing factors?N) rr9r6uniqueranger4rxrytupler) r factor_levelswi cell_countrXkeyr error_messagecounts rrzAnovaRM._check_data_balancedsG ++ 9B S2!5!5!78 8M 9 499??1-. $EC{{ 7 499S>..u56 7*Cj ",S/A"5 3"# 3 $. z?m +]+ +3 0C 3' // 0 99??1  5 5)* * 6rc  |j|jj}|jDcgc]}d|z }}d|jz}||gz}t j dj||j}|jj}|D]H}tjdg|jdz} d| ||<tj| ||<Jdj|g} t|| |jd} |dd| f}t||} | j} | j |jdkr t#d | D]}|j%||D] }||| ||<| j&} | j(}| j*}gd }t-j.tj0d | }|D]}|j|vs|d k7st3|||| |g\}}||z }||z |z }|dj|ddk(s |dz|z|vr||z }|}n&t3|||| |dz|zg\}}||z }||z |z }||z }t4j6j9|||}|j;ddj;dd}||j<|df<||j<|df<||j<|df<||j<|df<t?|Scc}w)zvestimate the model and compute the Anova table Returns ------- AnovaResults instance z C(%s, Sum)*r6Fr!T:Nz$Independent variables are collinear.)F ValueNum DFDen DFPr > F)rr.r& InterceptrzC(z, Sum)rrrr) r6rrrrpatsydmatrixjoinr7term_name_slicesr;rVr4rr fitrankrpoprr]r\rrr<rrr^r_replacer[ AnovaResults)rrrfrrrorrrr term_excluderresultsrr]r\r' anova_tablessr1 df_resid1df1msmmsedf2rprs rrz AnovaRM.fit&s IIdkk " ) )-1KK8q,"88-G9$ MM#((7+$)) <mm44  -C((E71771:-.C$(C C !!xx}K  -)* lAGGAJ? afIAq ))+ :: "CD D A OOA   5C*3/4K  5##kk;ll288F#3WE  4C||3&3++="4q+vu#6i(*czS(388GCRL11sW,K?.C"C&81k6sW,-'/OD)$h.C#:,C#IGGJJq#s+{{4,44XrB34 i025 h/25 h/23 h// 42K((m9s L)NNN)__name__ __module__ __qualname____doc__rrrrrrrrs(5nDH $$<0 *B@)rrc"eZdZdZdZdZdZy)rzX Anova results class Attributes ---------- anova_table : DataFrame c||_yN)r)rrs rrzAnovaResults.__init__qs &rc>|jjSr)summary__str__rs rrzAnovaResults.__str__ts||~%%''rctj}|jd|j|j|S)zlcreate summary results Returns ------- summary : summary2.Summary instance Anova)r Summary add_titleadd_dfr)rsumms rrzAnovaResults.summaryws5! w D$$% rN)rrrrrrrrrrrris'( rr__main__)olsz moore.csvr!)partner_status conformity fcategoryfscore)skiprowsrGz5conformity ~ C(fcategory, Sum)*C(partner_status, Sum)rz#conformity ~ C(partner_status, Sum)r*)r )&statsmodels.compat.pythonrnumpyr;pandasrrrrrtr statsmodels.formula.formulatoolsrrr statsmodels.iolibr #statsmodels.regression.linear_modelr rrIr=r>r?rrrrrrstatsmodels.formula.apirread_csvmoorermoore_lmmooreBrHrrrr s,#  '3 F"9>x4nVp%NgTD})})@8 z+ FOOK!#9 :EJ # 6U C G G IF X1 %E/r