{smcl} {* 28jun2015}{...} {* @@ Written by Elliott Lowy, mostly on the US government's dime (17 US Code ยง 105).}{...} INCLUDE help also_vlowy {title:Title} {pstd}{bf:models} {hline 2} Summarize estimation results {title:Syntax} {phang2}{cmd:models clear} {phang2}{cmd:models add} [{it:model label}] {phang2}{cmd:models label} {it:integer} [{it:model label}] {phang2}{cmd:models} {opt di:splay} [{cmd:,} {it:display options}] {phang2}{cmd:models all} [{it:model label}] [{cmd:,} {it:display options} ] {title:Description} {pstd}This suite of commands is for saving estimation results and displaying a summary table. {phang}o-{space 2}{cmd:models clear} clears any existing saved results and prepares for {cmd:models add}.{p_end} {phang}o-{space 2}{cmd:models add} adds the {bf:current} estimation results, as well as current name- and value- labels, to the set of saved results. {phang}o-{space 2}{cmd:models label} lets you retrospectively set or remove the {it:model label} from any of the results. {it:integer} references the models by the order in which they were added, starting at {cmd:1}. {phang}o-{space 2}{cmd:models} {opt di:splay} displays the table of summarized results.{p_end} {phang}o-{space 2}{cmd:models all} is a shortcut for the other {cmd:models} sub-commands (in order): {cmd:models clear}, {cmd:models add}, {cmd:models display}. {title:Display Options} {phang}{opt inc:lude(result types)} allows you to specify the display for each variable in a model. The allowed {it:result types} are {cmd:beta}, {cmd:p}, and {cmd:ci}[{cmd:(}{it:percent}{cmd:)}]. {phang2}o-{space 2}When {opt inc:lude()} is not specified, {cmd:beta} and {cmd:p} are used.{p_end} {phang2}o-{space 2}When {cmd:ci} is specified alone, a 95% confidence interval is used.{p_end} {phang2}o-{space 2}{cmd:beta} actually displays the exponentiated coefficients, with the appropriate label.{p_end} INCLUDE help tabel_options2n {pmore}{it:nl1} governs the IVs, and {it:nl2} governs the DVs. INCLUDE help tabel_options2v INCLUDE help tabel_out2 {title:Remarks} {pstd}Types of models currently supported are: linear, logistic, poisson, cox, and competing risk. Dummy variable notation ({cmd:i.}) is working, but interaction notation ({cmd:#}) isn't, yet. {pstd}I believe it should be easy at this point to add more types of models or results. Let me know if something more would be useful...