Identification strategy

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

Identification strategy

Post by jdavalos » Fri Sep 06, 2019 10:45 am

Congrats for the very interesting proposal,

Among your instruments, the 1st and 2nd risk to be a function of Firm performance:

- HH assets are a function of the HH entrepreneurial skills while microfinance funding at the many regions is attracted by the firms’ potential to generate profits.

Interest rate may also be higher to price in higher credit risk in regions where firms are less profitable.

This is of course an a priori assessment. Emprically, the availability of more than 1 instrument will allow you to implement an over identification test (Hansen) that will shed light on the validity of your proposed instruments.

Alternatively, you can argue that identification could be achieved through the non linearities implied by the binary endogenous regressor and its probit specification. A similar argument has been developed in the literature in the context of the identification of Inverse Mills Ratios in Heckman (2 Step) WITHOUT exclusion restriction. You can borrow the same argument from the following references

http://www2.gsu.edu/~ecort/jae2286.pdf
See footnote (10) 

Leung S, Yu S. 2000. Collinearity and two-step estimation of sample selection models: problems, origins, and remedies. Computational Economics 15: 173–199

Puhani P. 2000. The Heckman correction for sample selection and its critique. Journal of Economic Surveys 14: 53–68.

A suggestion would be to provide alternative estimations to assess the robustness of your findings and have a sense of the alternative identification strategies (without instruments and identification through non-linearities, with instruments+identification through non linearities)

Post Reply

Return to “PMMA-20396 - Accessing finance for productive rural employment opportunities for women in Benin”

Who is online

Users browsing this forum: No registered users and 1 guest