# table of results #1 stargazer(parallel, se=list(cse(parallel)), title="RD regression",type="text", column.labels=c("parallel"), df=FALSE,digits=2)

## ## RD regression ## =============================================== ## Dependent variable: ## --------------------------- ## All ## parallel ## ----------------------------------------------- ## D 7.66*** ## (1.51) ## ## agec -0.97 ## (0.66) ## ## Constant 91.84*** ## (0.71) ## ## ----------------------------------------------- ## Observations 48 ## R2 0.59 ## Adjusted R2 0.58 ## Residual Std. Error 2.49 ## F Statistic 32.99*** ## =============================================== ## Note: *p<0.1; **p<0.05; ***p<0.01

# table of results #2 stargazer(parallel, linear, quad, cubic, se=list(cse(parallel),cse(linear),cse(quad),cse(cubic)), title="RD regressions",type="text", column.labels=c("parallel","linear","quadratic","cubic"), df=FALSE,digits=2)

## ## RD regressions ## ======================================================== ## Dependent variable: ## ------------------------------------ ## All ## parallel linear quadraticcubic ## (1) (2) (3) (4) ## -------------------------------------------------------- ## D 7.66*** 7.66*** 9.55***11.26*** ## (1.51) (1.27) (1.83) (2.53) ## ## agec -0.97 0.83 -0.83 -5.63 ## (0.66) (0.72) (2.85) (8.05) ## ## I(agec2) -0.84 -6.92 ## (1.54) (9.81) ## ## I(agec3) -2.06 ## (3.25) ## ## D:agec -3.60*** -6.02 -6.79 ## (1.12) (4.53) (11.26) ## ## D:I(agec2) 2.90 16.05 ## (2.26) (12.87) ## ## D:I(agec3) -0.33 ## (4.16) ## ## Constant 91.84*** 93.62*** 93.07***92.28*** ## (0.71) (0.63) (0.78) (1.48) ## ## -------------------------------------------------------- ## Observations 48 48 48 48 ## R2 0.59 0.67 0.68 0.69 ## Adjusted R2 0.58 0.65 0.64 0.63 ## Residual Std. Error 2.49 2.28 2.29 2.32 ## F Statistic 32.99*** 29.47*** 18.02***12.67*** ## ======================================================== ## Note: *p<0.1; **p<0.05; ***p<0.01

1. How do we know the RDD in result #1 is a centered RD model?
2. Why does Bill call the model in result #1 "parallel?"
3. What is the advantage of a centered RD model?
4. What is the name of our treatment variable from the code in result #1?
5. What do our results from the RDD in result #2 say about the effect of legal access to alcohol on deaths?
6. Interpret and discuss the impact of the coefficient on D:agec in the second column of the table of results produced using result #2