74723_en-US

liliconcom

Missing value handling for Ordinal x/Factors in Partition

Suggestion: Please treat missing data (and missing value code data) for ordinal X/Factor variables the same as continuous X/Factor variables when the informative missing option is selected. That is, test to see if missing is better grouped "high" or "low". Right now, it appears to be treat the missing value (or missing value code) as an ordinal value and assign it accordingly.

Statistics and AnalyticsRe: Missing value handling for Ordinal x/Factors in Partition - Status changed to: In Development

We're working on this for JMP 16. 

 

In the meantime, a way to force ordering is to treat missing in the Value Ordering property.

 

Column("age")<Re: Missing value handling for Ordinal x/Factors in Partition

Tracking Number is 7612567893

Re: Missing value handling for Ordinal x/Factors in Partition

Hi -- 

 

I performed a quick retest on this in 16.  It appears to work when the dependent variable is continuous, but when the dependent variable is ordinal, it appears to just put the missing value category of the ordinal predictor variable with the low category group in the split.  So, I think this is still an open wish.  Thanks.

Re: Missing value handling for Ordinal x/Factors in Partition

Here's a JSL snippet to reproduce what we're seeing...

bigclass = Open("$SAMPLE_DATA/Big Class.jmp");
bigclass << New Column("MissingLow",Modeling Type(Ordinal),<< Formula(If( :age < 14 & Random Integer( 2 ) - 1,.,:age)));
bigclass << New Column("MissingHigh",Modeling Type(Ordinal),<< Formula(If( :age > 14 & Random Integer( 2 ) - 1,.,:age)));
bigclass << Partition(
 Y( :age ),
 X( :MissingLow ), // replace age with missing for some 12 and 13 year olds
 Informative Missing( 1 ),
 Split Best( 2 )
); // Note first split puts (missing, 12, 13) into a group
bigclass << Partition(
 Y( :age ),
 X( :MissingHigh ), // replace age with missing for some 15-17 year olds
 Informative Missing( 1 ),
 Split Best( 2 )
); // Note no split puts missing into a group with any 15, 16, or 17 year olds, it puts missing on the low end only

Re: Missing value handling for Ordinal x/Factors in Partition

@SamGardner it looks like this is coming, but not in JMP 16 or 17 is that what you see?  Should the status actually be investigated?