Econ B2000, MA Econometrics

Kevin R Foster, the Colin Powell School at the City College of New York, CUNY

Fall 2019

For this lab, we use interactions in regression models to (as usual!) explain wages.

Form a group of 3. Groups should prepare a 4-min presentation by one of the group members about their experiment process and results. You get 45 min to prepare.

Build on previous labs in creating useful models.

In this case we’ll take on a question (suggested by Sarah Marrara and Ramona Hernandez of CUNY’s Dominican Studies Institute ), about whether Dominicans see lower returns to schooling. You can refine by comparing with other Hispanic groups (eg Mexican) and/or checking how this changes for immigrants or children of immigrants.

First off, want to specify who we count as Dominican, for instance:

dominican_heritage <- (acs2017_ny$ANCESTR1 == "Dominican") | (acs2017_ny$ANCESTR2 == "Dominican") | (acs2017_ny$BPLD == "Dominican Republic")

Then identify a subgroup (for example, prime age fulltime workers). Later you might also check if Dominicans are more or less likely to be in some of those categories – what is labor force participation rate for Dominican men or women relative to other groups? What is rate of fulltime or full year work?

Try a regression with a Dominican dummy variable and interactions of this dummy with all of the education dummies. What do you find? Does this change if you have various polynomial specifications for Age? What if you interact those age profiles with Dominican dummy? What if you use \(log(Wage)\)?

What are the other variables you are using in your regression? Do they have the expected signs and patterns of significance? Explain if there is a plausible causal link from X variables to Y and not the reverse. Explain your results, giving details about the estimation, some predicted values, and providing any relevant graphics. Impress.

If you find this interesting, it’s an active research topic and you can work with DSI researchers to investigate further.