Exercise

# Practice Computing Causal Effects Using Two-Stage Least Squares (2SLS): CreditCo

Now let us repeat the previous CreditCo analysis, but now use 2SLS to compute the causal effect instead of indirect inference.

Instructions

**100 XP**

- 1) Using a linear regression, compute the first-stage effect (i.e. correlation between opting in and the rain storm)
- 2) Compute the predicted values of opting in
- 3) Compute the causal effect by linear regression (i.e. correlation between credit balance and predicted opt-in rate)
- 4) Did opting in to the credit offer increase or decrease credit balances?