data Fitness1;
input Oxygen RunTime @@;
datalines;
44.609 .
54.297 .
49.874 .
. .
39.442 .
50.541 .
44.754 .
;
run;
proc mi data=Fitness1 seed=1518971 simple nimpute=0;
em itprint outem=outem;
var Oxygen RunTime;
run;
here log says: WARNING: All observations are missing for variable RunTime. This variable will be excluded from the analysis.
I want to ask if a variable is all missing, the EM algorithm can't estimate the parameter? Why?
@diwang wrote:
data Fitness1; input Oxygen RunTime @@; datalines; 44.609 . 54.297 . 49.874 . . . 39.442 . 50.541 . 44.754 . ; run; proc mi data=Fitness1 seed=1518971 simple nimpute=0; em itprint outem=outem; var Oxygen RunTime; run;
here log says: WARNING: All observations are missing for variable RunTime. This variable will be excluded from the analysis.
I want to ask if a variable is all missing, the EM algorithm can't estimate the parameter? Why?
What do I have my pocket? (to quote a famous hobbit)
The procedure is attempting to make a "reasonable" estimate to replace missing values using information from records with non-missing values to build the rules. If there are no non-missing values then their is nothing to build a rule from.
Thanks for your answer.
But the EM algorithm can estimate the parameter with variables unobserved. Such as the Oxygen is all observed and RunTime is all unobserved. Here assume (Oxygen, RunTime ) ~ Normal (mu,sigma) ,and we use EM algorithm can estimate (mu,sigma). Is this wrong ?
Are you ready for the spotlight? We're accepting content ideas for SAS Innovate 2025 to be held May 6-9 in Orlando, FL. The call is open until September 25. Read more here about why you should contribute and what is in it for you!
Learn how use the CAT functions in SAS to join values from multiple variables into a single value.
Find more tutorials on the SAS Users YouTube channel.