Hello
I hope you have a good day
I would like to please answer the following questions
How can I include asymetric country-pair fixed effects in the "ppmlhdfe" command?
Previously I have built the so-called pairs of asymmetric countries and the pairs of symmetric countries with the following formula:
*Asimetric country-pair
egen pair_id_asimetricos=group(iso3_o iso3_d)
replace pair_id_asimetricos = pair_id_asimetricos +100000 if iso3_o==iso3_d
I have run the following regression:
ppmlhdfe tot_trade_millon ln_pib_o_ ln_pib_d_ ln_gdpcap_o ln_gdpcap_d ln_REM_EXP_1 ln_REM_IMP_1, absorb(pair_id_asimetricos) cluster(pair_id_asimetricos)
What is striking is the following:
Absorbed degrees of freedom:
-------------------------------------------------------------+
Absorbed FE | Categories - Redundant = Num. Coefs |
---------------------+---------------------------------------|
pair_id_asimetricos | 25652 25652 0 *|
-------------------------------------------------------------+
* = FE nested within cluster; treated as redundant for DoF computation
Why does it show me that my asimetrics pair fixed effects are redundant ?
why does it show me "num coef iqual to 0"
Thank you for yours answers
best wishes
I hope you have a good day
I would like to please answer the following questions
How can I include asymetric country-pair fixed effects in the "ppmlhdfe" command?
Previously I have built the so-called pairs of asymmetric countries and the pairs of symmetric countries with the following formula:
*Asimetric country-pair
egen pair_id_asimetricos=group(iso3_o iso3_d)
replace pair_id_asimetricos = pair_id_asimetricos +100000 if iso3_o==iso3_d
I have run the following regression:
ppmlhdfe tot_trade_millon ln_pib_o_ ln_pib_d_ ln_gdpcap_o ln_gdpcap_d ln_REM_EXP_1 ln_REM_IMP_1, absorb(pair_id_asimetricos) cluster(pair_id_asimetricos)
What is striking is the following:
Absorbed degrees of freedom:
-------------------------------------------------------------+
Absorbed FE | Categories - Redundant = Num. Coefs |
---------------------+---------------------------------------|
pair_id_asimetricos | 25652 25652 0 *|
-------------------------------------------------------------+
* = FE nested within cluster; treated as redundant for DoF computation
Why does it show me that my asimetrics pair fixed effects are redundant ?
why does it show me "num coef iqual to 0"
Thank you for yours answers
best wishes