Comments on your proposal

Post Reply
jdavalos
Posts: 21
Joined: Sun Jun 10, 2018 8:35 am
Contact:

Comments on your proposal

Post by jdavalos » Thu Sep 05, 2019 12:34 pm

Congratulations for the very interesting and well written proposal. My comments below:

- Consider referring to the original Dubin and McFadden (DMF) paper rather than Bourguigons'. The latter provides a slight contribution while their bootstrap study is indeed informative. DMF contribution is substantial.

- You consider your outcome (productivity variable) as a censored one (with respect to the alternative credit constraints categories). Why? This seems misleading. Endogenous credit constraints does not necessarily imply a complete new regime for the productivity outcome unless you assume heterogenous effects of the credit access. Do you mean to model heterogenous effects (by analogy to an endogenous switching regression model)?

- I would first consider a more parsimonious model that does not assume heterogenous effects. This implies modeling your outcomes (y) as a function of a multinomial credit constraint endogenous variable (M), the latter modeled by a multinomial logit with exclusion restrictions. This can be estimated with the GSEM stata module or the CMP stata package :
Y = f(X,M,E)
E = g(Z,X)
X are exogenous while Z represent your exclusion restriction (instruments).

- Should you be able to motivate the heterogenous effects argument, then you should implement the Dubin and McFadden (or Bourguignon) specification.

Congrats once again

Post Reply

Return to “PMMA-20443 - Credit constraints, agricultural productivity and household welfare in Burkina Faso: a gender perspective”

Who is online

Users browsing this forum: No registered users and 1 guest