p7]c @sddlZddlmZddlZddlmZmZddlZddl m Z ddl m Z ddl mZmZmZddlmZdZd Zd Zd Zd Zd ZdZdZdefdYZdefdYZedkrddlZddlm Z ej!ddddddddgZ"e dde"j#Z$e d de"j#Z%ee$d!d"Z&ndS(#iN(tstats(t DataFrametIndex(tOLS(tlrange(t_remove_intercept_patsyt_has_interceptt_intercept_idx(tsummary2cCs|dkr|jS|dkr2|j}|jS|dkrN|j}|jS|dkrj|j}|jS|dkr|j}|j St d|dS(Nthc0thc1thc2thc3s robust options %s not understood( tNonet cov_paramstHC0_setcov_HC0tHC1_setcov_HC1tHC2_setcov_HC2tHC3_setcov_HC3t ValueError(tmodeltrobusttse((s6lib/python2.7/site-packages/statsmodels/stats/anova.pyt_get_covariance s          c Ks|jdd}|jdd}|jdd}|jdd}|r]|j}n|jj}|jj}|jd}|jj} |jjj } |jj } t | j t | d} d|} d d d || g}ttj| d fd |}|dkr7t||||| || || | S|dkr\t|| | || |S|dkrt|| | || |S|dkrtdntdt|dS(s9 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. ttesttFtscalettypiRisPR(>%s)tdftsum_sqtmean_sqitcolumnstIitIIitIIIitIVsType IV not yet implementedsType %s not understoodN(iR$(iR%(iR&(iR'(tgetR tlowerRtendogtexogtshapet endog_namestdatat design_infot exog_namestlenttermsRRtnptzerostanova1_lm_singletanova2_lm_singletanova3_lm_singletNotImplementedErrorRtstr(RtkwargsRRRRR*R+tnobst response_nameR/R0tn_rowstpr_testtnamesttable((s6lib/python2.7/site-packages/statsmodels/stats/anova.pyt anova_single!s8      !    c Cs%t|dd } | d krNtjj|\} } tj| j|} ntjt|j t|j f} g|j D]}|j |^q}x*t |D]\}}d| ||fRRBtqtrtarrtnametslicestitslice_R!tidxRJRO((s6lib/python2.7/site-packages/statsmodels/stats/anova.pyR5]s2 '%    4% (cCs|j}t|}dd||g}ttj|dfd|}t|d } t||} g} g} xat|D]S\} }|j|}t |j |j }g}t |j }x|D]{}t |j }|j|r||k r|j|}|jt |j |j |jt |j |j qqWtj|jjjd|}tj|jjjd|}|jr!tjtj|| |j}ddlm}|j|\}}|jd|jd}tj|d d | d fj|}n|}|jd}|d kr|j|d | }|j|j|j| |f<}|j|j|j| |fRt terms_infoR?R@tcovt robust_covt col_orderROR]ttermtcolstL1tL2tterm_setttt other_settcoltLVLREt orth_complt_RYtL12RUt test_valueRS((s6lib/python2.7/site-packages/statsmodels/stats/anova.pyR6sX  ! #   !/  ! .  +cCs|t|7}|j}dd||g}ttj|dfd|}t||} g} g} xt|D]\} } |j| }tj|j j j d|}|}|j d}|dkr$|j |d| }|j |j|j| |f<}|j|j|j| |fRRpR?R@RqRsROR]RtRuRvRRYRURRS((s6lib/python2.7/site-packages/statsmodels/stats/anova.pyR7s2 !   !   +c Os|jdd}t|dkr;|d}t||S|dkr`tdt|n|jdd}|jdd}t|}d |}d d d d ||g}ttj|dfd|} |s|dj }ng|D]} | j ^q| d 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) RiiR$s6Multiple models only supported for type I. Got type %sRRRsPr(>%s)RTRStdf_difftss_diffiR#i(iR$N(R(R1RARR9R RR3R4RRSRTtdifftvaluesRRRORRURVRWtisnull( targsR:RRRRtn_modelsR>R?R@tmdl((s6lib/python2.7/site-packages/statsmodels/stats/anova.pytanova_lms2F     !  + cCs?tjtg|}x"|D]}||}t|| FiiR#t InterceptisC(ts, Sum)(ii("R.RRRRtpatsytdmatrixtjoinR/tterm_name_slicesR3RMRR,RRRtfittrankRtpopRRTRStpdRR4RRRURVtreplaceRRt AnovaResults(RRR]RRRdRRRRt term_excludeRtresultsRRTRSR#t anova_tabletssr1t df_resid1tdf1tmsmtmsetdf2RtpRt((s6lib/python2.7/site-packages/statsmodels/stats/anova.pyR'sb   !               N(t__name__t __module__t__doc__R RRRR(((s6lib/python2.7/site-packages/statsmodels/stats/anova.pyRs 7   "RcBs)eZdZdZdZdZRS(sX Anova results class Attributes ---------- anova_table : DataFrame cCs ||_dS(N(R(RR((s6lib/python2.7/site-packages/statsmodels/stats/anova.pyRsscCs|jjS(N(tsummaryt__str__(R((s6lib/python2.7/site-packages/statsmodels/stats/anova.pyRvscCs-tj}|jd|j|j|S(smcreate summary results Returns ------- summary : summary2.Summary instance tAnova(RtSummaryt add_titletadd_dfR(Rtsumm((s6lib/python2.7/site-packages/statsmodels/stats/anova.pyRys  (RRRRRR(((s6lib/python2.7/site-packages/statsmodels/stats/anova.pyRks  t__main__(tolss moore.csvtskiprowsiR?tpartner_statust conformityt fcategorytfscores5conformity ~ C(fcategory, Sum)*C(partner_status, Sum)R.s#conformity ~ C(partner_status, Sum)Ri('tnumpyR3RiRtpandasRRRRt#statsmodels.regression.linear_modelRtstatsmodels.compat.pythonRt statsmodels.formula.formulatoolsRRRtstatsmodels.iolibRRRAR5R6R7RRRtobjectRRRtstatsmodels.formula.apiRtread_csvtmooreRtmoore_lmtmooreBR@(((s6lib/python2.7/site-packages/statsmodels/stats/anova.pyts8     < 7 X ' j  #