Problem using xtabond2

[Dani Resar]

Hello everyone,

I am trying to use xtabond2 for a regression, but I keep getting the error: "Equation not identified. Regessors outnumber instruments."

I keep getting this error even when I am using formulations in the published literature, for example:
xtabond2 ghegdp_who_s l.ghegdp_who_s govdisgdp_up_net pridisbgdp_up_net drdisgdp gdppc_usd_imf ggegdp_wb hiv_prevalence, /// gmm(l.ghegdp_who_s l.govdisgdp_up_net) iv(govdisgdp_up_net pridisbgdp_up_net drdisgdp gdppc_usd_imf ggegdp_wb hiv_prevalence, eq(diff)) robust h(2)
I have run this same code using my own variables below:

xtabond2 ghes_gdp l.ghes_gdp dahg_gdp dahng_gdp debt_gdp log_gdp_cap_v2 gge_gdp hiv_prev
gmm(l.ghe_gdp l.dah_gdp) iv(dah_gdp dahng_gdp debt_gdp log_gdp_cap_v2 gge_gdp hiv_prev, eq(diff)) robust h(2)

and get the error described above. This is not in fact the regression I am using for my analysis, I am using a variation of this, but my variation is not running either and I thought it would be simplest to ask this question in context of the published regression as they were clearly able to overcome whatever error I am running into.

Thanks
Danielle

[Sebastian Kripfganz]

There is likely to be a problem with time series gaps in your data set. Did you maybe create five-year averages but forgot to declare the delta(5) option with xtset? Without more knowledge about your data set, it is not possible to provide any help.