Hello,
I'm trying to run 'xtmixed' model with autoregressive of 1 (AR1) using a propensity score matched sample. I tried to run the following code and received an error message:
xtmixed y x##z [fweight=weight] || group: || id: x, residual(ar 1, t(z))
The error message I got was "obervational-level weights not allowed with correlated residual structures.'
First of all, I had to use frequency weights to adjust the nature of control units selected by matching process (nearest neighborhood matching with n=5 and replacement). In addition, I wanted to address the potential correlation between y_t-1 and y_t. Is there any way I could fix this issue? Thank you for your concerns and help in advance!
I'm trying to run 'xtmixed' model with autoregressive of 1 (AR1) using a propensity score matched sample. I tried to run the following code and received an error message:
xtmixed y x##z [fweight=weight] || group: || id: x, residual(ar 1, t(z))
The error message I got was "obervational-level weights not allowed with correlated residual structures.'
First of all, I had to use frequency weights to adjust the nature of control units selected by matching process (nearest neighborhood matching with n=5 and replacement). In addition, I wanted to address the potential correlation between y_t-1 and y_t. Is there any way I could fix this issue? Thank you for your concerns and help in advance!
