I'm working on time-series data and want to convert multiple rows to column. I have tried using transpose and crosstab, but seem to me i'm doing something wrong. See below for sample record.
See output record below
Solved! Go to Solution.
Can you attach the data instead of screenshots? Typically you'll want to use a sample tool to isolate the top 4 rows, transpose those, and join them back to your dataset.
Hi @hay_001
To accomplish what you're trying to do, you must apply two different treatments to your input file.
First 4 rows need different operations than the remaining ones.
I'd split the file (maybe using a RecordID and filtering the first four rows and the remaining ones).
Then start transposing & CrossTabing.
At the end, merge the two datastreams to get the desired output.
Hi @hay_001
As mentioned above, isolate the headers and then join back to the data. The top piece gets the 5 header rows and flips them to be columns. Then it joins back to the dates and values.
@Luke_C, Awesome and thank you, that worked..