clear	
	clear matrix
	set matsize 800
	set mem 500m
	cd [your directory partition that has data files]
	
********
*Load the  efip, FG, and BBR team prs measures

*********

	
use [efip_bbr_fg_prs_measures.csv as .dta] 





*******
* add team fielding-runs saved data
*******

merge m:1 teamID yearID using [fielding_rs.csv], nogen

	
**************************
*create per ip variables
**************************
	
	gen rpip=RA // team runs-allowed per ip
	

	
	bys yearID: egen z_rpip=std(rpip)
	
	
	gen efiprs_ip=efip_prs/ip
        gen bbrar_ip=bbr_rar/ip
        gen fgrar_ip=fg_rar/ip



	
************************
*season standardization
************************
	
	bysort yearID: egen z_eprsip=std(efiprs_ip)
        bysort yearID: egen z_fruns=std(f_runs)
	bysort yearID: egen z_fgrarip=std(fgrar_ip)
        bysort yearID: egen z_bbrarip=std(bbrar_ip)





	
	
	
		
		

		




		
*****************************
* 1900-2025 regression models
*****************************	
					
		
		regress z_rpip z_eprsip z_fruns   [aweight=ip]
	
		regress z_rpip  z_fgrarip z_fruns   [aweight=ip]
		
		regress z_rpip z_bbrarip  z_fruns   [aweight=ip]
		
		regress z_rpip z_bbraaip z_fruns   [aweight=ip]
		regress z_rpip    z_fruns [aweight=ip]
	
		
*****************************************************
* multi-season regression models 
*****************************************************
* calculate incremental R^2s
gen Rbase = .
gen Ri_bbr = .
gen Ri_fg = .
gen Ri_efip= .

foreach yr of numlist 1900/2025 {
// Run the regression with base only for the current year
regress rpip f_runs if year == `yr'
// Store R2 from the first model  for the current year

local r2_1 = e(r2)
replace Rbase = `r2_1' if year == `yr'
// Run the regression with base and fg_rar for the current year
regress   rpip f_runs fgrar_ip if year == `yr' 
// Store incremental FG R2 for the current year
local r2_2 = e(r2)
replace Ri_fg = `r2_2'-Rbase if year == `yr'
// Run the regression with base and bbr for the current year
regress   rpip f_runs bbrar_ip if year == `yr' 
// Store incemental BBR R2 the current year
local r2_3 = e(r2)
replace Ri_bbr = `r2_3'-Rbase if year == `yr'
// Run the regression with base & efip for the current year
regress rpip f_runs efiprs_ip if year == `yr'
// Store incemental efip R2 from the third model for the current year
local r2_4 = e(r2)
replace Ri_efip = `r2_4'-Rbase if year == `yr'
di `yr'
}


preserve
collapse (mean) Rbase Ri_bbr Ri_fg Ri_efip,by(yearID)
drop if year ==2020
* Create cumulative variables for ribbon placement
gen cum_base = Rbase*100
gen cum_fg = (cum_base) + Ri_fg*100
gen cum_bbr = (cum_base) + Ri_bbr*100
gen cum_efip = (cum_base) + Ri_efip*100

* Smooth the data for each cumulative variable
foreach lname in base fg bbr efip {
lpoly cum_`lname' year,bw(2) gen(smooth_`lname') at(year)
}
* Create a variable for the lower bound

gen zero = 0
twoway ///
rarea smooth_base zero yearID, color(gs12%60) alcolor(gs12) || /// Smoothed Base
rarea smooth_bbr smooth_base yearID, color(blue%30) alcolor(blue%100) || /// Smoothed Increment
rarea  smooth_fg smooth_base yearID, color(green%30) alcolor(green%100) || ///
rarea smooth_efip smooth_base yearID, color(red%30) alcolor(red%100) ||  /// Smoothed Increment
, legend(off) ///
title(" ") ///
xlabel(1900(10)2025, labsize(large) nogrid angle(45)format(%ty)) ///
ylabel(0(25)100, labsize(large) nogrid angle(0)) ///
ytitle("") ///
xtitle("")
	
	
	restore