<p>下面是个program，只需注意加粗部分</p><p>program define skewt, rclass<br/>version 10.1<br/>syntax varlist (numeric)[if] [in], [bs saving(string) size(integer 4) reps(integer 1000) replace *]<br/>marksample touse <br/>foreach var of varlist `varlist' { <br/>capture confirm numeric variable `var'<br/> if _rc==0 {<br/>  qui sum `var' if `touse', detail<br/>  di ""<br/>  di in gr _col(5)  "`var'- stats from the sample"<br/>  di ""<br/>  di in gr _col(5) <strong>"N coefficient  = `=sqrt(r(N))'"</strong><br/>  di in gr _col(5) "S-coefficient  = `=r(mean)/r(sd)'"<br/>  di in gr _col(5) "G-coefficient  = `r(skewness)'"<br/>  di in gr _col(5) "Sample mean    = `r(mean)'"<br/>  di ""<br/>  local ratio = (`=sqrt(r(N))')*((`=r(mean)/r(sd)') + ((1/3) * (`r(skewness)') * ((`=r(mean)/r(sd)')^2)) + ((1/(6*((`=sqrt(r(N))')^2)))* (`r(skewness)')))<br/>  return scalar ratio_`var'=`ratio'<br/> }<br/> else {<br/>  di as input "`var'" as text " is not a numeric variable skewness adjusted t-statistic cannot be calculated."<br/> }</p><p>if "`bs'" == "bs" {<br/> if "`saving'" != "" {<br/>  local saving `saving'_`var'<br/> }<br/> bootstrap r(ratio_`var'), saving("`saving'", `replace') reps(`reps') size(`=int(_N/`size')') `options': skewt `var'<br/> }<br/>}<br/>end<br/></p><p>在上述命令中，加粗的部分不理想，因为只能返回no of observation，不管missing还是nonmissing，我想返回no of nonmissing observation，怎么达到，谢谢</p>