Hi, this is my first post in Statalist and I'm hoping you can help me.

I am using a panel data set with really good coverage 2001-2018, however, the DTI for 1999 and 2000 is shaky at best for a specific group (where DTI drops to 50 obs where there should be 10,000.) We used multiple imputations as a fix but when compared to a benchmark, they fell 7 points lower than expected DTI's.

I have a benchmark for the 10th, 25th, 50th, 75th, and 90th percentiles from a reliable source for 1999 and 2000 with a total count for the year.

I have never created weights before/ have barely used them.
But should I make sample weights for my specific group for 2000, or create buckets?

I know this is vague, but I'm not sure what information would help for clarity.

Best and Thanks,
Sian