Instrumental Variables in R exercises (Part-3)

May 29, 2017
By

(This article was first published on R-exercises, and kindly contributed to R-bloggers)

This is the third part of the series on Instrumental Variables. For other parts of the series follow the tag instrumental variables.

In this exercise set we will use Generalized Method of Moments (GMM) estimation technique using the examples from part-1 and part-2.
Recall that GMM estimation relies on the relevant moment conditions. For OLS we assume that predictors are uncorrelated with the error term. Similarly, IV estimation implies that the instrument(s) is uncorrelated with the error term.

Answers to the exercises are available here.

Exercise 1
Load the AER package (package description: here) and the PSID1976 dataset. This has data regarding labor force participation of married women.
Define a new data-frame that has data for all married women that were employed. As we did in part-2, this data-frame will be used for the remaining exercises.
Next, load the ‘gmm’ package (package description: here).

Exercise 2
We will start with a simple example. Regress log(wage) on education using the usual OLS technique.
Next, use the gmm function to estimate the same model using OLS’s moment conditions. Match your result and comment on the standard errors.

Exercise 3
Estimate the return to education for the model from Exercise-2 using feducation as the IV. Use both ivreg and gmm functions and compare results.

Exercise 4
Regress log(wage) on education, experience and experience^2 using the usual OLS technique.
Next, use the gmm function to estimate the same model using OLS’s moment conditions. Match your result.

Exercise 5
Estimate the return to education for the model from Exercise-4 using feducation as the IV. Use both ivreg and gmm functions and compare results.

Exercise 6
We will now use the over-identified case. Estimate the return to education for the model from Exercise-2 using feducation and meducation as IVs. Use both ivreg and gmm functions and compare results.

Exercise 7
Estimate the return to education for the model from Exercise-4 using feducation and meducation as IVs. Use both ivreg and gmm functions and compare results.

Exercise 8
The test of over-identifying restrictions can be obtained by the J-test (Sargan test). It is displayed with the summary and specTest functions. Do the over-identified moment conditions match the data well?

Exercise 9
Iterated estimation might offer some advantages over the default two-step method in some cases. Estimate the model in Exercise-7 using the iterative estimation technique.

Exercise 10
Use the plot function to get the graph of log(wage) and fitted values for the model in Exercise-7.

To leave a comment for the author, please follow the link and comment on their blog: R-exercises.

R-bloggers.com offers daily e-mail updates about R news and tutorials on topics such as: Data science, Big Data, R jobs, visualization (ggplot2, Boxplots, maps, animation), programming (RStudio, Sweave, LaTeX, SQL, Eclipse, git, hadoop, Web Scraping) statistics (regression, PCA, time series, trading) and more...



If you got this far, why not subscribe for updates from the site? Choose your flavor: e-mail, twitter, RSS, or facebook...

Comments are closed.

Search R-bloggers


Sponsors

Never miss an update!
Subscribe to R-bloggers to receive
e-mails with the latest R posts.
(You will not see this message again.)

Click here to close (This popup will not appear again)