{smcl} {* 31May2009}{...} {cmd:help rmse} {hline} {title:Title} {p2colset 5 13 22 2}{...} {p2col :{hi: rmse} {hline 2}}Caluclate raw or adjusted RMSE (Root MSE){p_end} {marker s_Syntax} {title:Syntax} {p 4 4 6} {cmdab:rmse} {it:variable} {it:variable} [{it:{help varlist}}] [, {it:options}] {marker s_Description} {title:Description} {p 4 4 6} {cmd:rmse} provides a fast and easy way to calculate raw or adjusted RMSE, useful for checking prediction accuracy. {p 4 4 6}rmse = [(sum of (y-yhat)^2)/(N - k - 1)]^.5{p_end} {p 4 4 6}where k is the number of parameters, usually regression coefficients.{p_end} {p 4 4 6} The results are stored in r-class locals. {marker s_Options} {title:Options} {dlgtab:Main} {p 4 12 6}{opt est:imates} Use the same sample and degrees of freedom (df) as the current regression results.{p_end} {p 4 12 6}{opt df_m(#)} # is k in the formula above. Comparable to e(df_m){p_end} {p 4 12 6}{opt raw} Do not reduce the df by 1 [as in at the end of (N - k - 1)]{p_end} {marker s_0} {title:Examples} {p 4 4 6}* reproduce a single RMSE from a regression{p_end} {p 4 4 6}{stata sysuse auto, clear}{p_end} {p 4 4 6}{stata regress mpg price rep78 headroom turn}{p_end} {p 4 4 6}{stata predict yhat}{p_end} {p 4 4 6}{stata di e(rmse)}{p_end} {p 4 4 6}{stata rmse mpg yhat, est}{p_end} {p 4 4 6}* reproduce from scratch{p_end} {p 4 4 6}{stata sysuse auto, clear}{p_end} {p 4 4 6}{stata regress mpg price rep78 headroom}{p_end} {p 4 4 6}{stata predict yhat}{p_end} {p 4 4 6}{stata gen resid=yhat-mpg}{p_end} {p 4 4 6}{stata gen resid2=resid*resid}{p_end} {p 4 4 6}{stata sum resid2 if e(sample)==1}{p_end} {p 4 4 6}{stata di e(rmse)}{p_end} {p 4 4 6}{stata di (r(mean)*e(N)/(e(N)-e(df_m)-1))^.5}{p_end} {p 4 4 6}{stata rmse mpg yhat if e(sample)==1, df_m(3)}{p_end} {p 4 4 6}* multiple RMSE{p_end} {p 4 4 6}{stata sysuse auto, clear}{p_end} {p 4 4 6}{stata rmse mpg price rep78 headroom}{p_end} {p 4 4 6}{stata ret list}{p_end} {title:Remarks} {p 4 12 6}The first variable is compared against the rest.{p_end} {title:Author} {p 4 4 6}Roy Wada{p_end} {p 4 4 6}roywada@hotmail.com{p_end}