Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • GMM: Multiple lagged moment conditions


    Hello everyone! I am a complete newbie to Stata and am thus really grateful for any help. I am currently trying to implement a highly non-linear model in Stata using GMM. I am struggling to understand how to

    1) Incorporate lagged moment conditions
    2) Implement this highly non-linear regression.

    My moment conditions should be:
    E[price_s * error_t] = 0 for s < t, t = 2,3,...,T

    The form of the error term is highly non-linear, i.e.
    error_t = demand_t - ((1 - {theta} * z_t) ({alpha} * + {beta} * controlvar + {delta} AvPrice_t) + {theta} * z_(t-1) * ({alpha} * + {beta} * controlvar + {delta} AvPrice_t)

    My biggest issue right now is the fact that I only have a single moment condition and do not now how to incorporate the lagged condition mentioned above. I have looked at other non-linear GMM specifications, but failed to understand how they obtain multiple moment conditions. I know that my lack of understanding runs deep and would appreciate either some help on my question or someone pointing me to a resource that I can use.

    Thank you very much in advance! (and if there is any other information you need, I am sorry for not providing it and will try to do so asap).
Working...
X