ó áp7]c @s®dZddlmZddlmZddlZddljZ e d„Z d„Z d„Z dZd Zd „Zd „Zd „Zed krªdZejjddedfƒjdefdefdefdefgƒZejjdedfƒjdefdefgƒZddlZejjj eefde!ƒZ"ej#edfe"ƒZ$x"ej%j&D]Z'ee'e$e'Z?e1e?j@ƒe1e?j%ƒed+e?ƒ\ZAZBej)ged,eBƒD]Z*eAe*^q%ƒZCe?d&ZDe j.eDeCƒj/ƒZEe1eEj2ƒe1eeeEƒƒd-jFe5j%j&d ƒZGed.e?ƒ\ZHZIej)ged,eIƒD]Z*eHe*^qŃZJe?d&ZKe j.eKeJƒj/ƒZLe1eLj2ƒe1eeeLƒƒx†eID]{ZMe1d/eMfƒej)geeMeIƒD]Z*eHe*^qTƒZNe?d&ZOe j.eOeNƒj/ƒZPe1eeePƒƒq(WndS(0s convenience functions for ANOVA type analysis with OLS Note: statistical results of ANOVA are not checked, OLS is checked but not whether the reported results are the ones used in ANOVA includes form2design for creating dummy variables TODO: * ... * iÿÿÿÿ(tprint_function(tlmapNcCs|jƒ}tj|ƒ}|rD|dd…df|kjtƒS|dd…df|kjtƒdd…dd…fSdS(s|convert array of categories to dummy variables by default drops dummy variable for last category uses ravel, 1d onlyNiÿÿÿÿ(traveltnptuniquetNonetastypetint(txt returnalltgroups((sKlib/python2.7/site-packages/statsmodels/sandbox/regression/try_ols_anova.pyt data2dummys  #cCsltjtt|jƒƒƒ}||dd…ddd…fkjdƒjjt ƒdd…dd…fS(sÝcreates product dummy variables from 2 columns of 2d array drops last dummy variable, but not from each category singular with simple dummy variable but not with constant quickly written, no safeguards Niÿÿÿÿ( RRRttuplettolistRtalltTRR(RR ((sKlib/python2.7/site-packages/statsmodels/sandbox/regression/try_ols_anova.pytdata2proddummy s cCsB|jdkr(|dd…df}nt|dtƒ}||S(sýcreate dummy continuous variable Parameters ---------- x1 : 1d array label or group array x2 : 1d array (float) continuous variable Notes ----- useful for group specific slope coefficients in regression iNR (tndimRR tTrue(tx1tx2tdummy((sKlib/python2.7/site-packages/statsmodels/sandbox/regression/try_ols_anova.pytdata2groupcont/ssW ANOVA statistics (model sum of squares excludes constant) Source DF Sum Squares Mean Square F Value Pr > F Model %(df_model)i %(ess)f %(mse_model)f %(fvalue)f %(f_pvalue)f Error %(df_resid)i %(ssr)f %(mse_resid)f CTotal %(nobs)i %(uncentered_tss)f %(mse_total)f R squared %(rsquared)f s] ANOVA statistics (model sum of squares includes constant) Source DF Sum Squares Mean Square F Value Pr > F Model %(df_model)i %(ssmwithmean)f %(mse_model)f %(fvalue)f %(f_pvalue)f Error %(df_resid)i %(ssr)f %(mse_resid)f CTotal %(nobs)i %(uncentered_tss)f %(mse_total)f R squared %(rsquared)f c Cs‰i}|j|jƒddddddddd d d g }x!|D]}t||ƒ||>> xx, n = form2design('I a F:b P:c*d G:c*f', testdata) >>> xx.keys() ['a', 'b', 'const', 'cf', 'cd'] >>> n ['const', 'a', 'b', 'cd', 'cf'] Notes ----- with sorted dict, separate name list wouldn't be necessary tIitconstt:isF:isP:t*tsG:sunknown expression in formula( tsplitRtonestshapetappendR Rtc_tjoinRt ValueError(tsstdatatvarstnamestitemtv((sKlib/python2.7/site-packages/statsmodels/sandbox/regression/try_ols_anova.pyt form2designms.$  6,cCs/|}x!|jƒD]}|j|ƒqW|S(swdrop names from a list of strings, names to drop are in space delimeted list does not change original list (R2tremove(R9tlitnewliR=((sKlib/python2.7/site-packages/statsmodels/sandbox/regression/try_ols_anova.pytdropnameªst__main__ièitsizeitatbtctditetftflattenis I a F:b P:c*dsI a F:b P:c*d G:a*e fg{®Gáz„?sae ftbreedtsextlittertpentpigtagetbagetys dftest3.datatmissingt.tusemasks I F:sex ageR1t s'I F:breed F:sex F:litter F:pen age bages Results dropping(Qt__doc__t __future__Rtstatsmodels.compat.pythonRtnumpyRtstatsmodels.apitapitsmtFalseR RRt anova_str0t anova_strR,R?RCt__name__R"trandomtrandinttviewRt testdatainttnormaltfloatt testdataconttnumpy.lib.recfunctionstlibt recfunctionst zip_descrRtdt2temptyttestdatatdtypeR<tnametxxt column_stacktnntXtsumRTtOLStfittrest1tprinttparamstdt_bt genfromtxttdtatktmasktbooltmtreshapetlentanytdroprowsR:t dta_use_b1R4txx_b1tnames_b1tX_b1ty_b1trest_b1R7tallexogtxx_b1at names_b1atX_b1aty_b1atrest_b1atdropntX_b1a_ty_b1a_t rest_b1a_(((sKlib/python2.7/site-packages/statsmodels/sandbox/regression/try_ols_anova.pyt s€      = K6 !&& /&  !3'  /  /   /