!g# dZddlmZddlZddlmZd1dZdZ dZ dZ dZ d Z d Zd Zed k(rd Zej$j'dedfj)defdefdefdefgZej$j/edfj)defdefgZddlZej6j8j;eefdZej>edfeZ ejBjDD] Z#ee#e e#< ejBjDD] Z#ee#e e#<  ede \Z$Z"ede \Z$Z"ejPe"Dcgc]}e$| c}Z)e)jUddej$j/ezzZ+ejXe+e)j[Z.e&e.j^e&e e e.zejPede"Dcgc]}e$| c}Z)e)jUddej$j/ezzZ+ejXe+e)j[Z.e&e.j^e&e e e.zejBdefd efd!efd"efd#efd$efd%efd&efgZ0ejbd'e0d(d)Z2e&d*e2jBjDDcgc]}e2jf|jU!c}e2jfj)e4Z5e5jmd+e7e2jBjDjqdZ9e2e9ddfjtZ;e&e;jxe&e;jBed,e;\Z=Z>ejPed-e>Dcgc]}e=| c}Z?e;d&Z@ejXe@e?j[ZAe&eAj^e&e e eAzd.je2jBjDdd+ZCed/e;\ZDZEejPed-eEDcgc]}eD| c}ZFe;d&ZGejXeGeFj[ZHe&eHj^e&e e eHzeED]mZIe&d0eIfejPeeIeEDcgc]}eD| c}ZJe;d&ZKejXeKeJj[ZLe&e e eLzoyycc}wcc}wcc}wcc}wcc}wcc}w)2a convenience functions for ANOVA type analysis with OLS Note: statistical results of ANOVA are not checked, OLS is checked but not whether the reported results are the ones used in ANOVA includes form2design for creating dummy variables TODO: * ... * )lmapNc|j}tj|}|r|dddf|k(jtS|dddf|k(jtddddfS)z|convert array of categories to dummy variables by default drops dummy variable for last category uses ravel, 1d onlyN)ravelnpuniqueastypeint)x returnallgroupss i/var/www/dash_apps/app1/venv/lib/python3.12/site-packages/statsmodels/sandbox/regression/try_ols_anova.py data2dummyrsh  A YYq\F!T' f$,,S11!T' f$,,S1!CRC%88ctjtt|j }||dddddfk(j dj jtddddfS)zcreates 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 Nr) rrrtupletolistallTr r )r r s rdata2proddummyrs^YYtE188:. /F vaQh  $ $R ( * * 1 1# 6q"u ==rcV|jdk(r |dddf}t|d}||zS)zcreate 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 NT)r )ndimr)x1x2dummys rdata2groupcontr.s2 ww!| $Z rT *E 2:raW ANOVA statistics (model sum of squares excludes constant) Source DF Sum Squares Mean Square F Value Pr > F Model %(df_model)i %(ess)f %(mse_model)f %(fvalue)f %(f_pvalue)f Error %(df_resid)i %(ssr)f %(mse_resid)f CTotal %(nobs)i %(uncentered_tss)f %(mse_total)f R squared %(rsquared)f a] ANOVA statistics (model sum of squares includes constant) Source DF Sum Squares Mean Square F Value Pr > F Model %(df_model)i %(ssmwithmean)f %(mse_model)f %(fvalue)f %(f_pvalue)f Error %(df_resid)i %(ssr)f %(mse_resid)f CTotal %(nobs)i %(uncentered_tss)f %(mse_total)f R squared %(rsquared)f ci}|j|jgd}|D]}t||||<|jj|d<|j |j z |d<|S)zjupdate regression results dictionary with ANOVA specific statistics not checked for completeness ) df_modeldf_residessssruncentered_tss mse_model mse_resid mse_totalfvaluef_pvaluersquarednobs ssmwithmean)update__dict__getattrmodelr*r#r")resad anova_attrkeys r anovadictr4[sp BIIcllJ$#s#3$BvJ**SWW4B} Irci}g}|jD]}|dk(r7tj|jd|d<|j d@d|vr||||<|j |^|dddk(r7|jdd}t ||||<|j ||ddd k(r|jddjd }t tj||d||df|d j|<|j d j|&|ddd k(ro|jddjd }t||d||d|d j|<|j d j|td ||fS)aconvert string formula to data dictionary ss : str * I : add constant * varname : for simple varnames data is used as is * F:varname : create dummy variables for factor varname * P:varname1*varname2 : create product dummy variables for varnames * G:varname1*varname2 : create product between factor and continuous variable data : dict or structured array data set, access of variables by name as in dictionaries Returns ------- vars : dictionary dictionary of variables with converted dummy variables names : list list of names, product (P:) and grouped continuous variables (G:) have name by joining individual names sorted according to input Examples -------- >>> 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 would not be necessary Irconst:NzF:rzP:*zG:zunknown expression in formula) splitronesshapeappendrrc_joinr ValueError)ssdatavarsnamesitemvs r form2designrIlsH D E > 3;GGDJJqM2DM LL ! _dDJ LL  "1X  3"A a)DG LLO "1X  3"((-A-beeD1JtAaDz4I.JKD  LL $ "1X  3"((-A-d1Q4j$qt*ED  LL $<= =)>* ;rc\|dd}|jD]}|j||S)zwdrop names from a list of strings, names to drop are in space delimited list does not change original list N)r<remove)rClinewlirGs rdropnamerNs3 qEE  T Lr__main__i)sizeabcdr9efT)flattenrz I a F:b P:c*dzI a F:b P:c*d G:a*e fg{Gz?zae fbreedsexlitterpenpigagebageyz dftest3.data.)missingusemaskrcrz I F:sex ager; z'I F:breed F:sex F:litter F:pen age bagez Results dropping)F)M__doc__statsmodels.compat.pythonrnumpyrstatsmodels.apiapismrrr anova_str0 anova_strr4rIrN__name__r*randomrandintviewr testdataintnormalfloat testdatacontnumpy.lib.recfunctionslib recfunctions zip_descrdt2emptytestdatadtyperFnamexxnprintr@ column_stackXsumraOLSfitrest1paramsdt_b genfromtxtdtamaskboolmreshapelenanydroprowsrD dta_use_b1r>xx_b1names_b1X_b1y_b1rest_b1rAallexogxx_b1a names_b1aX_b1ay_b1arest_b1adropnX_b1a_y_b1a_ rest_b1a_)nnks00rrs> + 9 >0   ";z z D))##AT!H#5::SIs3iQTUXPY[^_bZc;deK99##4(#3883u+E{9STL! )) * *K+Ft * TCrxxa#&H!!''+$T*+""((,%d+,  OX6IB3X>IB %0BB01A a4 ((t(555A BFF1QKOO E %,, )i& &' (65*ABBBBCA a4 ((t(555A BFF1QKOO E %,, )i& &' 288gs^eS\Hc?3<%u~E?S%L2 3D "--S$ GC )AAchhqkoo'AB  dAyyC 0155a8HhYq[!&&J *   *   "-