missing valures imputation

[abdelilah arredouani]

Hi all,
I am trying to impute my missing values but i i get an error message i can't figure out what is wrong. the summarize command gives me the table below:



Variable | Obs Mean Std. Dev. Min Max
-------------+---------------------------------------------------------

bodyfatbmi | 500 29.07252 5.741849 15.9 53.9
salivasali~w | 500 81.89784 55.9216 1 .31 276.71
plasmasali~l | 478 30.8422 12.9376 3.19 72.42
glucoseact | 494 5.742591 2.098976 3.7 19.1


i can see missing values in plasmasalivaryamyalseul glucoseact variables, which i register for imputation.
then i run the code below but i get amessage saying:

glucoseact: missing imputed values produced
This may occur when imputation variables are used as independent variables or when independent variables contain
missing values. You can specify option force if you wish to proceed anyway

What I am doing wrong?


Code:
mi set wide mi register imputed plasmasalivaryamyalseul glucoseact mi describe mi impute chained (regress) plasmasalivaryamyalseul glucoseact = bodyfatbmi salivasalivaryamylaseumlnew, add (20) rseed(20) note: missing-value pattern is monotone; no iteration performed

[Carlo Lazzaro]

Abdelilah:
have you already taken a look at -mi impute monotone- entry in Stata .pdf manual?

[abdelilah arredouani]

problem solved

[Carlo Lazzaro]

Abdelilah:
other listers might be interested in the fix you found out.
Why not sharing the diagnosis of your problem and the fix you applied? Thanks.

[abdelilah arredouani]

Hi carlo,
i was about to do it and i found your post. in fact what happened is that when i copied my data from excel to the data editor the cells below the cells with the values of my variables contained a dot and stata, for some reason, considered them as missing values. Once i deleted them the prole was solved.

regards

Stata/IC 15.1 for Mac

[Carlo Lazzaro]

Thanks, Abdelilah.
Your experience shows that even a trivial mishap can stop a statistical anlysis.
In my opinion, checking dataset consistency, especially when importing in Stata from external sources/softwares, can never be overestimated.