# Talk:Inverse Mills ratio

WikiProject Statistics (Rated Start-class, Low-importance)

This redirect is within the scope of the WikiProject Statistics, a collaborative effort to improve the coverage of statistics on Wikipedia. If you would like to participate, please visit the project page or join the discussion.

Start  This redirect does not require a rating on the quality scale.
Low  This redirect has been rated as Low-importance on the importance scale.

## Possible mistake (turned out not to be a mistake)

I think you have a problem in the conditional mean formula. The corrected formula should be

{\displaystyle {\begin{aligned}&\operatorname {E} [\,x\,|\ x>\alpha \,]=\mu +\sigma ^{2}{\frac {\phi {\big (}{\tfrac {\alpha -\mu }{\sigma }}{\big )}}{1-\Phi {\big (}{\tfrac {\alpha -\mu }{\sigma }}{\big )}}},\\&\operatorname {E} [\,x\,|\ x<\alpha \,]=\mu +\sigma ^{2}{\frac {-\phi {\big (}{\tfrac {\alpha -\mu }{\sigma }}{\big )}}{\Phi {\big (}{\tfrac {\alpha -\mu }{\sigma }}{\big )}}},\end{aligned}}}

I have double checked by integration (using Mathematica).--132.64.128.115 (talk) 15:01, 27 October 2011 (UTC)

A later addition (by the same user): this was a mistake. If, instead of using the CDF and PDF of standard normal distribution and standardized bound one simply uses the relevant normal distribution with proper mu and sigma (for example, using "normpdf(x,mu,sigma)" in MATLAB), then one has to multiply the ratio by an additional sigma, simply because PDF(x,mu,sigma)=sigma*phi((x-mu)/sigma). But the equation are right. I am leaving this point because the version history tells me that this is not the first time someone makes this error.--192.114.91.245 (talk) 16:11, 16 November 2011 (UTC)