help estaddalso see:esttab,estout,eststo,estposthttp://repec.org/bocode/e/estout -------------------------------------------------------------------------------Titleestadd-- Add results to (stored) estimatesSyntaxestaddsubcommand[,options] [:namelist] wherenamelistis_all|*|name[name...]subcommandsdescription ----------------------------------------------------------------- Elementarylocalname ...add a macroscalarname=expadd a scalarmatrixname=matadd a matrixr(name)add contents ofr(name)(matrix or scalar) Statistics for each coefficientbetastandardized coefficientsvifvariance inflation factors (afterregress)pcorrpartial (and semi-partial) correlationsexpbexponentiated coefficientsebsdstandardized factor change coefficientsmeanmeans of regressorssdstandard deviations of regressorssummvarious descriptives of the regressors Summary statisticscoxsnellCox & Snell's pseudo R-squarednagelkerkeNagelkerke's pseudo R-squaredlrtestmodellikelihood-ratio testysummdescriptives of the dependent variable Othermarginsadd results frommargins(Stata 11 or newer) SPostbrantadd results frombrant(if installed)fitstatadd results fromfitstat(if installed)listcoefadd results fromlistcoef(if installed)mlogtestadd results frommlogtest(if installed)prchangeadd results fromprchange(if installed)prvalueadd results fromprvalue(if installed)asprvalueadd results fromasprvalue(if installed) -----------------------------------------------------------------optionsdescription -----------------------------------------------------------------replacepermit overwriting existinge()'sprefix(string)specify prefix for names of added resultsquietlysuppress output from subcommand (if any)subcmdoptssubcommand specific options -----------------------------------------------------------------Descriptionestaddadds additional results to thee()-returns of an estimation command (see help estcom, helpereturn). If nonamelistis provided, then the results are added to the currently active estimates (i.e. the model fit last). If these estimates have been previously stored, the stored copy of the estimates will also be modified. Alternatively, ifnamelistis provided after the colon, results are added to all indicated sets of stored estimates (see helpestimates storeor helpeststo). You may use the*and?wildcards innamelist. Execution is silent ifnamelistis provided. Adding additional results to thee()-returns is useful, for example, if the estimates be tabulated by commands such asestoutoresttab. See the Examples section below for illustration of the usage ofestadd. Technical note: Some of the subcommands below make use of the information contained ine(sample)to determine estimation sample. These subcommands return error if the estimates do not containe(sample).+------------+ ----+ Elementary +-------------------------------------------------------Subcommandsestaddlocalname ...adds in macroe(name)the specified contents (also see helpereturn).estaddscalarname=expadds in scalare(name)the evaluation ofexp(also see helpereturn).estaddscalarr(name)adds in scalare(name)the value of scalarr(name).estaddscalarnameadds in scalare(name)the the value of scalarname.estaddmatrixname=matrix_expressionadds in matrixe(name)the evaluation ofmatrix_expression(also see helpmatrix define).estaddmatrixr(name)adds in matrixe(name)a copy of matrixr(name).estaddmatrixnameadds in matrixe(name)a copy of matrixname.estaddr(name)adds ine(name)the value of scalarr(name)or a copy of matrixr(name), depending on the nature ofr(name). +---------------------------------+ ----+ Statistics for each coefficient +----------------------------------estaddbetaadds ine(beta)the standardized beta coefficients.estaddvif[,tolerancesqrvif] adds ine(vif)the variance inflation factors (VIFs) for the regressors (see helpvif). Note thatvifonly works with estimates produced byregress.toleranceadditionally adds the tolerances (1/VIF) ine(tolerance).sqrvifadditionally adds the square roots of the VIFs ine(sqrvif).estaddpcorr[, semi] adds the partial correlations (see helppcorr) and, optionally, the semi-partial correlations between the dependent variable and the individual regressors (see, e.g., thepcorr2package from the SSC Archive). In the case of multiple-equations models, the results are computed for the first equation only. The partial correlations will be returned ine(pcorr)and, ifsemiis specified, the semi-partial correlations will be returned ine(spcorr).estaddexpb[,noconstant] adds ine(expb)the exponentiated coefficients (see the helpeform_option).noconstantexcludes the constant from the added results.estaddebsdadds ine(ebsd)the standardized factor change coefficients, i.e. exp(b_jS_j), where b_j is the raw coefficient and S_j is the standard deviation of regressor j, that are sometimes reported for logistic regression (see Long 1997).estaddmeanadds ine(mean)the means of the regressors.estaddsd[,nobinary] adds ine(sd)the standard deviations of the regressors.nobinarysuppresses the computation of the standard deviation for 0/1 variables.estaddsumm[,stats] adds vectors of the regressors' descriptive statistics to the estimates. The followingstatsare available:statsdescription -----------------------------------------------------------meanmeansumsumminminimummaxmaximumrangerange = max - minsdstandard deviationvarvariancecvcoefficient of variation (sd/mean)semeanstandard error of mean = sd/sqrt(n)skewnessskewnesskurtosiskurtosisp11st percentilep55th percentilep1010th percentilep2525th percentilep5050th percentilep7575th percentilep9090th percentilep9595th percentilep9999th percentileiqrinterquartile range = p75 - p25allall of the abovemedianequivalent to specifying "p50"qequivalent to specifying "p25 p50 p75" ----------------------------------------------------------- The default ismean sd min max. Alternatively, indicate the desired statistics. For example, to add information on the regressors' skewness and kurtosis, type . estadd summ, skewness kurtosis The statistics names are used as the names for the returnede()matrices. For example,estadd summ, meanwill store the means of the regressors ine(mean). +--------------------+ ----+ Summary statistics +-----------------------------------------------estaddcoxsnelladds ine(coxsnell)the Cox & Snell pseudo R-squared, which is defined as r2_coxsnell = 1 - ( L0 / L1 )^(2/N) where L0 is the likelihood of the model without regressors, L1 the likelihood of the full model, and N is the sample size.estaddnagelkerkeadds ine(nagelkerke)the Nagelkerke pseudo R-squared (or Cragg & Uhler pseudo R-squared), which is defined as r2_nagelkerke = r2_coxsnell / (1 - L0^(2/N))estaddlrtestmodel[,name(string)lrtest_options] adds the results from a likelihood-ratio test, wheremodelis the comparison model (see helplrtest). Added aree(lrtest_chi2),e(lrtest_df), ande(lrtest_p). The names may be modified using thename()option. Specifyname(myname)to adde(mynamechi2),e(mynamedf), ande(mynamep). See helplrtestfor thelrtest_options.estaddysumm[,stats] adds descriptive statistics of the dependent variable. See thesummsubcommand above for a list of the availablestats. The default ismean sd min max. The default prefix for the names of the added scalars isy(e.g. the mean of the dependent variable will be returned ine(ymean)). Useestadd'sprefix()option to change the prefix. If a model has multiple dependent variables, results for the first variable will be added. +-------+ ----+ Other +------------------------------------------------------------estaddmargins[marginlist] [if] [in] [weight] [,options] adds results from themarginscommand, which was introduced in Stata 11. See helpmarginsfor options. All results returned bymarginsexcepte(V)are added using "margins_" as a default prefix. For example, the margins are added ine(margins_b). The standard errors are added ine(margins_se). Use theprefix()option to change the default prefix. +-------+ ----+ SPost +------------------------------------------------------------ The following subcommands are wrappers for commands from Long and Freese'sSPostpackage (see http://www.indiana.edu/~jslsoc/spost.htm). Type . net from http://www.indiana.edu/~jslsoc/stata to obtain the latestSPostversion (spost9_ado).SPostfor Stata 8 (spostado) is not supported. For examples on using the subcommands see http://repec.org/bocode/e/estout/spost.html.estadd brant[,brant_options] appliesbrantfrom Long and Freese'sSPostpackage and adds the returned results toe(). You may specifybrant_optionsas described in helpbrant. The following results are added:e(...)Contents ------------------------------------------------------------ Scalarsbrant_chi2Chi-squared of overall Brant testbrant_dfDegrees of freedom of overall Brant testbrant_pP-value of overall Brant test MatrixbrantTest results for individual regressors (rows: chi2, p<chi2) ------------------------------------------------------------ To address the rows ofe(brant)inestout'scells()option typebrant[chi2]andbrant[p<chi2].estadd fitstat[,fitstat_options] appliesfitstatfrom Long and Freese'sSPostpackage and adds the returned scalars toe(). You may specifyfitstat_optionsas described in helpfitstat. Depending on model and options, a selection of the following scalar statistics is added:e(...)Contents ------------------------------------------------------------devDeviance (D)dev_dfDegrees of freedom of Dlrx2LR or Wald X2lrx2_dfDegrees of freedom of X2lrx2_pProb > LR or Wald X2r2_adjAdjusted R2r2_mfMcFadden's R2r2_mfadjMcFadden's Adj R2r2_mlML (Cox-Snell) R2r2_cuCragg-Uhler(Nagelkerke) R2r2_mzMcKelvey & Zavoina's R2r2_efEfron's R2v_ystarVariance of y*v_errorVariance of errorr2_ctCount R2r2_ctadjAdj Count R2aic0AICaic_nAIC*nbic0BICbic_pBIC'statabicBIC used by StatastataaicAIC used by Statan_rhsNumber of rhs variablesn_parmNumber of parameters ------------------------------------------------------------estadd listcoef[varlist] [,nosdlistcoef_options] applieslistcoeffrom Long and Freese'sSPostpackage and adds the returned results toe(). You may specifylistcoef_optionsas described in helplistcoef. Furthermore, optionnosdsuppresses adding the standard deviations of the variables ine(b_sdx). Depending on the estimation command and options, several of the following matrices are added:e(...)Contents ------------------------------------------------------------b_xsx-standardized coefficientsb_ysy-standardized coefficientsb_stdFully standardized coefficientsb_factFactor change coefficientsb_factsStandardized factor change coefficientsb_pctPercent change coefficientsb_pctsStandardized percent change coefficientsb_sdxStandard deviation of the Xs ------------------------------------------------------------ For nominal models (mlogit,mprobit) the original parametrization ofe(b)may not match the contrasts computed bylistcoef. To be able to tabulate standardized coefficients along with the raw coefficients for the requested contrasts, the following additional matrices are added for these models:e(...)Contents ------------------------------------------------------------b_rawraw coefficientsb_sestandard errors of raw coefficientsb_zz statisticsb_pp-values ------------------------------------------------------------estadd mlogtest[varlist] [,mlogtest_options] appliesmlogtestfrom Long and Freese'sSPostpackage and adds the returned results toe(). You may specifymlogtest_optionsas described in helpmlogtest. Depending on the specified options, a selection of the following returns are added:e(...)Contents ------------------------------------------------------------ Scalarshausman_set#_chi2Hausman IIA tests usinghausmanhausman_set#_dfhausman_set#_psuest_set#_chi2Hausman IIA tests usingsuestsuest_set#_dfsuest_set#_psmhsiao_set#_chi2Small-Hsiao IIA testssmhsiao_set#_dfsmhsiao_set#_pcombine_#1_#2_chi2Wald tests for combination of outcomescombine_#1_#2_dfcombine_#1_#2_plrcomb_#1_#2_chi2LR tests for combination of outcomeslrcomb_#1_#2_dflrcomb_#1_#2_pwald_set#_chi2Wald tests for sets of independentwald_set#_dfvariableswald_set#_plrtest_set#_chi2LR tests for sets of independentlrtest_set#_dfvariableslrtest_set#_pMatriceswaldWald tests for individual variables (rows: chi2, df, p)lrtestLR tests for individual variables (rows: chi2, df, p) ------------------------------------------------------------ To address the rows ofe(wald)ande(lrtest)inestout'scells()option type the row names in brackets, for example,wald[p]orlrtest[chi2].estadd prchange[varlist] [ifexp] [inrange] [,pattern(typepattern)binary(type)continuous(type)[no]avgsplit[(prefix)]prchange_options] appliesprchangefrom Long and Freese'sSPostpackage and adds the returned results toe(). You may specifyprchange_optionsas described in helpprchange. In particular, theoutcome()option may be used with models for count, ordered, or nominal outcomes to request results for a specific outcome. Further options are:pattern(typepattern),binary(type), andcontinuous(type)to determine which types of discrete change effects are added as the main results. The default is to add the 0 to 1 change effect for binary variables and the standard deviation change effect for continuous variables. Usebinary(type)andcontinuous(type)to change these defaults. Available types are:typeDescription ------------------------------------------------minmaxminimum to maximum change effect010 to 1 change effectdeltadelta()change effectsdstandard deviation change effectmargefctmarginal effect (some models only) ------------------------------------------------ Usepattern(typepattern)if you want to determine the type of the added effects individually for each regressor. For example,pattern(minmax sd delta)would addminmaxfor the first regressor,sdfor the second, anddeltafor the third, and then proceed using the defaults for the remaining variables.avgto request that only the average results over all outcomes are added if applied to ordered or nominal models (ologit,oprobit,slogit,mlogit,mprobit). The default is to add the average results as well as the individual results for the different outcomes (unlessprchange'soutcome()option is specified, in which case only results for the indicated outcome are added). Furthermore, specifynoavgto suppress the average results and only add the outcome-specific results.avgcannot be combined withsplitoroutcome().split[(prefix)] to save each outcome's results in a separate estimation set if applied to ordered or nominal models (ologit,oprobit,slogit,mlogit,mprobit). The estimation sets are namedprefix#, where#is the value of the outcome at hand. If noprefixis provided, the name of the estimation set followed by an underscore is used as the prefix. If the estimation set has no name (because it has not been stored yet) the name of the estimation command followed by an underscore is used as the prefix (e.g.ologit_). The estimation sets stored by thesplitoption are intended for tabulation only and should not be used with other post-estimation commands. Depending on model and options, several of the following matrices and scalars are added:e(...)Contents ------------------------------------------------------------ Scalarscentered1if effects are centered,0elsedeltaValue ofdelta()predval[#] Prediction(s) at the base valuesoutcomeOutcome value (outcome()/splitonly) MatricesdcDiscrete change effects (rows: main, minmax, 01, delta, sd [, margefct])patternTypes of effects in the main row ofe(dc)XBase values and descriptive statistics (rows: X, SD, Min, Max) ------------------------------------------------------------ Thee(dc)ande(X)matrices have multiple rows. Thee(dc)matrix contains the main results as determined bypattern(),binary(), andcontinuous()in the first row. The second and following rows contain the separate results for each type of effect using the labels provided byprchangeas row names. Typedc[#]ordc[rowname]to address the rows inestout'scells()option, where#is the row number orrownameis the row name. For example, typedc[-+sd/2]to address the centered standard deviation change effects. To tabulate the main results (1st row), simply typedc.e(pattern)indicates the types of effects contained in the main row ofe(dc)using numeric codes. The codes are 1 for the minimum to maximum change effect, 2 for the 0 to 1 change effect, 3 for thedelta()change effect, 4 for the standard deviation change effect, and 5 for the marginal effect.e(X)has four rows containing the base values, standard deviations, minimums, and maximums. If thefromtooption is specified, two additional matrices,e(dcfrom)ande(dcto)are added.estadd prvalue[ifexp] [inrange] [,label(string)prvalue_options]estadd prvaluepost[name] [,title(string)swap] appliesprvaluefrom Long and Freese'sSPostpackage and adds the returned results toe(). The procedure is to first collect a series of predictions by repeated calls toestadd prvalueand then applyestadd prvalue postto prepare the results for tabulation as in the following example: . logit lfp k5 k618 age wc hc lwg inc . estadd prvalue, x(inc 10) label(low inc) . estadd prvalue, x(inc 20) label(med inc) . estadd prvalue, x(inc 30) label(high inc) . estadd prvalue post . estout You may specifyprvalue_optionswithestadd prvalueas described in helpprvalue. For example, usex()andrest()to set the values of the independent variables. Uselabel()to label the single calls. "pred#" is used as label iflabel()is omitted, where # is the number of the call. Labels may contain spaces but they will be trimmed to a maximum length of 30 characters and some characters (:,.,") will be replaced by underscore. The results from the single calls are collected in matrixe(_estadd_prvalue)(predictions) and matrixe(_estadd_prvalue_x)(x-values). Specifyreplaceto drop results from previous calls.estadd prvalue postposts the collected predictions ine(b)so that they can be tabulated. The following results are saved:e(...)Contents ------------------------------------------------------------ ScalarsNnumber of observations Macrosdepvarname of dependent variablecmdestadd_prvaluemodelmodel estimation commandpropertiesbMatricesbpredictionssestandard errorsLBlower confidence interval boundsUBupper confidence interval boundsCategoryoutcome valuesCondconditional predictions (some models only)Xvalues of predictors (for each prediction)X2second equation predictors (some models only) ------------------------------------------------------------estadd prvalue postreplaces the current model unlessnameis specified, in which case the results are stored undernameand the model remains active. However, if the model has a name (because it has been stored), the name of the model is used as a prefix. If, for example, the model has been stored asmodel1, thenestadd prvaluepoststores its results undermodel1name. Usetitle()to specify a title for the stored results. The default forestadd prvalue postis to arrangee(b)in a way so that predictions are grouped by outcome (i.e. outcome labels are used as equations). Alternatively, specifyswapto group predictions byprvaluecalls (i.e. to use the prediction labels as equations).e(X)contains one row for each independent variable. To address the rows inestout'scells()option typeX[varname], wherevarnameis the name of the variable of interest.e(X2), if provided, is analogous toe(X).estadd asprvalue[,label(string)asprvalue_options]estadd asprvaluepost[name] [,title(string)swap] appliesasprvaluefrom Long and Freese'sSPostpackage and adds the returned results toe(). The procedure is to first collect a series of predictions by repeated calls toestadd asprvalueand then applyestadd asprvalue postto prepare the results for tabulation as in the following example: . clogit choice train bus time invc, group(id) . estadd asprvalue, cat(train bus) label(at means) . estadd asprvalue, cat(train bus) rest(asmean) label(at asmeans) . estadd asprvalue post . estout You may specifyasprvalue_optionswithestadd asprvalueas described in helpasprvalue. For example, usex()andrest()to set the values of the independent variables. Uselabel()to label the single calls. "pred#" is used as label iflabel()is omitted, where # is the number of the call. Labels may contain spaces but they will be trimmed to a maximum length of 30 characters and some characters (:,.,") will be replaced by underscore. The results from the single calls are collected in matricese(_estadd_asprval)(predictions),e(_estadd_asprval_asv)(values of alternative-specific variables), ande(_estadd_asprval_csv)(values of case-specific variables). Specifyreplaceto drop results from previous calls.estadd asprvalue postposts the collected predictions ine(b)so that they can be tabulated. The following results are saved:e(...)Contents ------------------------------------------------------------ ScalarsNnumber of observations Macrosdepvarname of dependent variablecmdestadd_asprvaluemodelmodel estimation commandpropertiesbMatricesbpredictionsasvalternative-specific variables (if available)csvcase-specific variables (if available) ------------------------------------------------------------estadd asprvalue postreplaces the current model unlessnameis specified, in which case the results are stored undernameand the model remains active. However, if the model has a name (because it has been stored), the name of the model is used as a prefix. If, for example, the model has been stored asmodel1, thenestadd asprvaluepoststores its results undermodel1name. Usetitle()to specify a title for the stored results. The default forestadd asprvalue postis to arrangee(b)in a way so that predictions are grouped by outcome (i.e. outcome labels are used as equations). Alternatively, specifyswapto group predictions byprvaluecalls (i.e. to use the prediction labels as equations).e(asv)ande(csv)contain one row for each variable. To address the rows inestout'scells()option typeasv[varname]orcsv[varname], wherevarnameis the name of the variable of interest.Optionsreplacepermitsestaddto overwrite existinge()macros, scalars, or matrices.prefix(string)denotes a prefix for the names of the added results. The default prefix is an empty string. For example, ifprefix(string)is specified, thebetasubcommand will return the matrixe(stringbeta).quietlysuppresses the output from the called subcommand and displays only the list of added results. Note that many ofestadd's subcommands do not generate output, in which casequietlyhas no effect.subcmdoptsare subcommand specific options. See the descriptions of the subcommands above.Example 1: AddExamplesr()-returns from other programs to the current estimates . sysuse auto (1978 Automobile Data) . quietly regress price mpg weight . test mpg=weight ( 1)mpg - weight = 0F( 1, 71) =0.36Prob > F =0.5514. estadd scalar p_diff = r(p) added scalar: e(p_diff) =.55138216. estout, stats(p_diff)------------------------- b ------------------------- mpg-49.51222weight1.746559_cons1946.069------------------------- p_diff.5513822------------------------- Example 2: Add means and standard deviations of the model's regressors to the current estimates . quietly logit foreign price mpg . estadd summ, mean sd added matrices: e(sd) :1 x 3e(mean) :1 x 3. estout, cells("mean sd") drop(_cons)-------------------------------------- mean sd -------------------------------------- price6165.257 2949.496mpg21.2973 5.785503-------------------------------------- Example 3: Add standardized beta coefficients to stored estimates . eststo: quietly regress price mpg (est1stored) . eststo: quietly regress price mpg foreign (est2stored) . estadd beta: * . estout, cells(beta) drop(_cons)-------------------------------------- est1 est2 beta beta -------------------------------------- mpg-.4685967 -.5770712foreign.2757378-------------------------------------- See http://repec.org/bocode/e/estout for additional examples.A program providing a newWriting one's own subcommandsestaddsubcommand should be calledestadd_mysubcommand(see helpprogramfor advice on defining programs).mysubcommandwill be available toestaddas a newsubcommandafter the program definition has been executed or saved to a file called "estadd_mysubcommand.ado" in either the current directory or somewhere else in theadopath(see helpsysdir). Use the subcommands provided within "estadd.ado" as a starting point for writing new subcommands. See http://repec.org/bocode/e/estout/estadd.html#estadd007 for an example.Ben Jann, Institute of Sociology, University of Bern, jann@soz.unibe.chAuthorManual:Also see[R] estimatesOnline: help forestimates,ereturn,program,esttab,estout,eststo,estpost