Export dot missings to Excel

[Zaneta]

Hi everyone,

I am trying to export some data to Excel from SAS EG 7.1 using proc export and dbms = xls. It works well except that one of my numeric variables includes both ordinary missings and .S and both are converted to blanks in the export. Is there a way to retain the .S as this or some other value that distinguishes it from the ordinary missings?

One solution is to use dbms = .csv but I then have problems with losing leading zeros in a character variable. And I also can't export multiple SAS datasets to different worksheets in the same Excel file then.

And another is to convert the numeric variable to a character - this changes .S to S which is fine, but converts the ordinary numeric missings to . in Excel plus all numbers to text which isn't so helpful.

Is there anything else I could try?

Many thanks,
Zaneta

[Kurt_Bremser]

In Excel, a missing numerical value is always represented as an empty cell. So if you want to have the column as numeric data, blanks is what you get.

A workaround could be to assign special non-missing values for your .S (before you do the export), which you could then use with a user-defined format in Excel, if that is possible.