*Create paths to folders 
clear all
global data "C:\Users\aapo.kivinen\Dropbox (Aalto)\PhD\teaching\Principles_of_Economic_Analysis\aapo\exercise 4\data"
global output "C:\Users\aapo.kivinen\Dropbox (Aalto)\PhD\teaching\Principles_of_Economic_Analysis\aapo\exercise 4\output"

ssc install xtable

*Import dataset to Stata
use "$data\CK1994.dta", replace 

*Label variables 
label define state 1 "New Jersey", add
label define state 0 "Pennsylvania", add
label values state state

label define time 0 "Pre", add
label define time 1 "Post", add
label values time time

*Generate variable for FTE 
gen fte = empft+0.5*emppt+nmgrs


preserve 

collapse (mean) wage_st fte hoursopen, by(time state)

sort state time
local variables wage_st fte 
foreach var of local variables {
	gen d_`var'_1 = `var'-`var'[_n-1] if state == state[_n-1]
	egen d_`var' = max(d_`var'_1), by(state)
	tabstat d*, by(state)
	gen diff_estimate_`var'_1 = d_`var'-d_`var'[_n-1] if state != state[_n-1]
	egen diff_estimate_`var' = max(diff_estimate_`var'_1)
}

*Q1.A and Q2.A
cd "C:\Users\aapo.kivinen\Dropbox (Aalto)\PhD\teaching\Principles_of_Economic_Analysis\aapo\exercise 4\output"
xtable  time state, c(mean fte mean d_fte_1) format(%9.3f) filename(Employment)
xtable  time state, c(mean wage_st mean d_wage_st_1) format(%9.3f) filename(Wage) 

drop *_1


*Q1.B and Q2.B
collapse (mean) diff_estimate_* 
export excel using "$output\Mean_Comparison_Results.xlsx", replace firstrow(variables)

restore 

*Q3 
gen NJ = 0 
replace NJ = 1 if state == 1

gen post = 0 
replace post = 1 if time == 1 

gen interaction = NJ*post 

reg fte NJ post interaction
eststo reg1
reg wage NJ post interaction
eststo reg2
esttab reg* using "regression_table.rtf", label star(* 0.1 ** 0.05 *** 0.01)  replace

*Q4 

keep if state == 1 
drop if wage_st == . 

gen dummy1 = 0 
replace dummy1 = 1 if wage_st<5 & time == 0 
egen dummy = max(dummy1), by(store)

label define dummy  0 "Over 5 $", add
label define dummy  1 "Under 5 $", add
label values dummy dummy

collapse (mean) wage_st fte hoursopen, by(time dummy)



sort dummy time
local variables wage_st fte 
foreach var of local variables {
gen d_`var'_1 = `var'-`var'[_n-1] if dummy == dummy[_n-1]
egen d_`var' = max(d_`var'_1), by(dummy)
tabstat d*, by(dummy)
gen diff_estimate_`var'_1 = d_`var'-d_`var'[_n-1] if dummy != dummy[_n-1]
egen diff_estimate_`var' = max(diff_estimate_`var'_1)
}

cd "C:\Users\aapo.kivinen\Dropbox (Aalto)\PhD\teaching\Principles_of_Economic_Analysis\aapo\exercise 4\output"
xtable  time dummy, c(mean fte mean d_fte_1) format(%9.3f) filename(Employment_NJ)
xtable  time dummy, c(mean wage_st mean d_wage_st_1) format(%9.3f) filename(Wage_NJ) 

drop *_1


collapse (mean) diff_estimate_* 
export excel using "$output\Mean_Comparison_Results_NJ.xlsx", replace firstrow(variables)

/*

use "$data\Commonpool_090614.dta", clear
keep kunta kuntaid merger year after free* mundebtpc mergerafter

regress mundebtpc merger after mergerafter, cluster(kuntaid)