{smcl} {* 14Feb2017/24Jun2013/05Jan2008/30dec2006/06sep2006/03aug2006}{...} {hline} help for {hi:fmlogit} {hline} {title:Fitting a fractional multinomial logit model by quasi maximum likelihood} {p 8 17 2} {cmd:fmlogit} {it:depvars} {weight} {ifin} [{cmd:,} {cmdab:eta:var(}{it:varlist}{cmd:)} {cmd:rpr} {cmdab:cl:uster(}{it:clustervar}{cmd:)} {cmdab:c:onstraints(}{it:numlist}|{it:matname}{cmd:})} {cmdab:l:evel(}{it:#}{cmd:)} {cmd:nolog} {cmdab:nocon:stant} {it:maximize_options} ] {p 4 4 2}{cmd:by} {it:...} {cmd::} may be used with {cmd:fmlogit}; see help {help by}. {p 4 4 2}{cmd:fweight}s, and {cmd:pweight}s are allowed; see help {help weights}. {p 4 4 2}{cmd:etavar} may contain factor variables; see {help fvvarlist}. {title:Description} {p 4 4 2} {cmd:fmlogit} fits by quasi maximum likelihood a fractional multinomial logit model. Each variable in depvarlist ranges between 0 and 1 and all variables in depvarlist must, for each observation, add up to 1: for example, they may be proportions. It is a multivariate generalization of the fractional logit model proposed by Papke and Wooldridge (1996). {p 4 4 2} Note that cases will be ignored if the one or more of the dependent variables has a value less zero or more than one or if the dependent variables don't add up to one. {p 4 4 2} Also note that {cmd:fmlogit} always implies the {help vce_option : vce(robust)} option because the model is fitted using quasi maximum likelihood. {title:Options} {p 4 8 2}{cmd:etavar()} specifies the explanatory variables. (The name of this option originates from the symbol commonly used for the linear predictor, the Greek letter eta.) {p 4 8 2}{cmd:rpr} reports the estimated coefficients transformed to relative proportion ratios, i.e., exp(b) rather than b. Standard errors and confidence intervals are similarly transformed. This option affects how results are displayed, not how they are estimated. {p 8 8 2}Relative proportion ratios can be useful when the model contains interaction terms, as in that case marginal effects as computed by dfmlogit will no longer be appropriate. Relative proportion ratios for the interaction terms can still be interpreted as the factor by which the relative proportion ratio changes, as is discussed in Buis (2010). {p 4 8 2}{cmd:cluster(}{it:clustervar}{cmd:)} specifies that the observations are independent across groups (clusters) but not necessarily within groups. {it:clustervar} specifies to which group each observation belongs; e.g., {cmd:cluster(personid)} in data with repeated observations on individuals. See {hi:[U] 23.14 Obtaining robust variance estimates}. {p 4 8 2} {cmdab:c:onstraints(}{it:numlist}|{it:matname}{cmd:})} specifies linear constraint(s) that are to be applied to the model; see help {help constraint}. {p 4 8 2}{cmd:level(}{it:#}{cmd:)} specifies the confidence level, in percent, for the confidence intervals of the coefficients; see help {help level}. {p 4 8 2}{cmd:nolog} suppresses the iteration log. {p 4 8 2}{cmdab:nocon:stant} suppresses the constant in the eta equations. {p 4 8 2}{it:maximize_options} control the maximization process; see help {help maximize}. If you are seeing many "(not concave)" messages in the log, using the {cmd:difficult} option may help convergence. {title:Example} {cmd} use http://fmwww.bc.edu/repec/bocode/c/citybudget.dta, clear gen pol = minorityleft + 2*noleft label define pol 0 "left parties are majority" /// 1 "left parties are minority" /// 2 "no left party" label value pol pol label var pol "political orientation of city government" fmlogit governing safety education recreation social urbanplanning, /// eta(i.pol houseval popdens) margins, dydx(*) predict(outcome(governing)) margins, dydx(*) predict(outcome(safety)) margins, dydx(*) predict(outcome(education)) margins, dydx(*) predict(outcome(recreation)) margins, dydx(*) predict(outcome(social)) margins, dydx(*) predict(outcome(urbanplanning)) {txt} {p 4 4 2}({stata "fmlogit_ex":click to run}){p_end} {title:Author} {p 4 4 2}Maarten L. Buis, University of Konstanz{break}maarten.buis@uni.kn {title:References} {p 4 4 2} Buis, M.L. 2010. Stata tip 87: Interpretation of interactions in non-linear models. {it:The Stata Journal} 10(2): 305-308. {p 4 4 2} Papke, Leslie E. and Jeffrey M. Wooldridge. 1996. Econometric Methods for Fractional Response Variables with an Application to 401(k) Plan Participation Rates. {it:Journal of Applied Econometrics} 11(6):619{c -}632. {title:Also see} {p 4 13 2} Online: help for {help fmlogit postestimation}, {p 4 13 2} If installed: {help dirifit}