clear matrix
	set matsize 800
	set mem 500m
	cd [your disk partition]
	use lahman pitching



	** back out hitter ABs
	*
drop if yearID <1900

	replace BAOpp = . if BAOpp == 9.99

	gen abf= H/BAOpp  

	drop if abf==.



	// sum data per pitcher



	collapse (sum) W L G GS CG SHO SV H ER HR BB SO  IBB WP HBP BK BFP GF R  GIDP IPouts abf, by(playerID yearID teamID lgID)


	// IP

	gen IP=IPouts/3





// import BIP data



	merge m:m playerID yearID using [1930_1979 BIP] ,update

	
	drop _merge
	
	
		merge m:m playerID yearID using [1980_1999 BIP],update

	
	drop _merge
	
	
	
	merge m:m playerID yearID using [1912_1929 BIP],update

	
	drop _merge
	
		
	
	
	
merge m:m playerID yearID using [2000-2024 BIP], update replace
	
	drop _merge

	
	drop if IP<100
	drop if IP==.



	
	
 replace teamID ="MIL" if teamID == "ML4" & year > 1997
 replace teamID ="ML4"  if teamID == "MIL" & year <1998
	replace teamID="LAA" if teamID =="ANA" & year >2004
	

merge m:m [BBR 2024 pitcher data],update 
drop _merge
drop if yearID==.

drop if IP==.
	


/// League id is not part of model but data retained for use in exploring alternative BIP_RBA models

gen AL=.
replace AL=1 if lgID=="AL"
replace AL=0 if lgID =="NL"
replace	AL=0	if teamID== 		"CHN"
replace	AL=0	if teamID== 		"WAS"
replace	AL=1	if teamID== 		"OAK"
replace	AL=0	if teamID== 		"SLN"
replace	AL=0	if teamID== 		"PIT"
replace	AL=1	if teamID== 		"LAA"
replace	AL=0	if teamID== 		"NYN"
replace	AL=1	if teamID== 		"MIN"
replace	AL=0	if teamID== 		"MIL"
replace	AL=1	if teamID== 		"HOU"
replace	AL=1	if teamID== 		"CLE"
replace	AL=0	if teamID== 		"PHI"
replace	AL=0	if teamID== 		"ATL"
replace	AL=1	if teamID== 		"TBA"
replace	AL=1	if teamID== 		"KCA"
replace	AL=1	if teamID== 		"DET"
replace	AL=1	if teamID== 		"SEA"
replace	AL=1	if teamID== 		"TEX"
replace	AL=1	if teamID== 		"NYA"
replace	AL=0	if teamID== 		"COL"
replace	AL=0	if teamID== 		"LAN"
replace	AL=1	if teamID== 		"BOS"
replace	AL=1	if teamID== 		"CHA"
replace	AL=0	if teamID== 		"SDN"
replace	AL=1	if teamID== 		"TOR"
replace	AL=1	if teamID== 		"BAL"
replace	AL=0	if teamID== 		"ARI"
replace	AL=0	if teamID== 		"SFN"
replace	AL=0	if teamID== 		"MIA"
replace	AL=0	if teamID== 		"CIN"


replace lgID ="AL" if AL==1 & lgID==""
replace lgID ="NL" if AL==0 & lgID==""


/// merge bbref SH SF
	merge m:m bbrefID yearID using [BBR SH/SF],update 
	drop _merge
	drop if yearID==.

replace abf=(H+(IP*3)) if year ==2024





drop if IP<100
drop if ifpop==.
drop if yearID==.
drop if SO ==.


*** determine performance values as rates (some vestigial stuff here)

gen gb_pct_rs2 =GB/(GB+FB+LD+POP)
gen k9=9*SO/IP
gen bb9=9*BB/IP
gen hr9=9*HR/IP
gen hr_pip=HR/IP
gen k_pip=SO/IP
gen bb_pip=BB/IP
gen hbp_pip=HBP/IP
drop if IP <100
gen rapg=9*R/IP
gen obabip=(H-HR)/(abf-(SO+HR)+SF)
gen gb_pct= gb/(gb+airb)
gen ifgb_pip=ifgb/IP
gen ifpop_pip=ifpop/IP
gen airb_pip=airb/IP
gen ofair_pip=ofair/IP
gen bobw= ifpop_pip-ofair_pip
gen gb_pct2= gb2/(gb2+airb2)
gen ifgb_pip2=ifgb2/IP
gen ifpop_pip2=ifpop2/IP
gen airb_pip2=airb2/IP
gen ofair_pip2=ofair2/IP
gen bobw2= ifpop_pip2-ofair_pip2
gen ifair_pip=ifair/IP
gen ifair_pip2=ifair/IP
regress rapg k_pip bb_pip hr_pip hbp_pip [iweight=IP]


/// calculate season-specific FIPr

 gen fipr=.
 
 foreach y of numlist 1912/2019 2021 2022 2023 2024 {
 	di "year `y'"
	regress rapg k_pip bb_pip hr_pip hbp_pip [iweight=IP] if year==`y'
	predict yhat 
replace fipr = yhat if year==`y'
drop yhat 
 }






/// merge team fielding scores

gen teamid = teamID


 merge m:m  teamid yearID using [team fielding data]
 
 drop _merge
 
 
 /// define BIP-era periods
 
gen period=.
 
 
 replace period = 1912 if yearID < 1923
 replace period= 1923 if yearID >= 1923 & yearID <1928
replace period = 1928 if yearID >= 1928 & yearID < 1935
replace period = 1935 if yearID >= 1935 & yearID < 1942
replace period = 1942 if yearID >= 1942 & yearID < 1955
replace period = 1955 if yearID >= 1955 & yearID < 1967
replace period = 1967 if yearID >= 1967 & yearID < 1975
replace period = 1975 if yearID >= 1975 & yearID < 1980
replace period = 1980 if yearID >= 1980 & yearID < 1990
replace period = 1990 if yearID >= 1990 & yearID < 1998
replace period = 1998 if yearID >= 1996 & yearID < 2002
replace period = 2002 if yearID >= 2002 & yearID < 2012
replace period = 2012 if yearID >= 2012 & yearID < 2017
replace period = 2017 if yearID >= 2017
 
 drop if period==.
 
 
 /// fit model with period-specific parameters 
 
 regress rapg i.period#c.(fipr field_composite ofair_pip ifpop_pip) [pweight=IP]
 
 
 /// save period parameters
gen cons=.

gen b_fipr = .
gen b_field =  .
gen b_ifpop =  .
gen b_ofair = .
 
 foreach p in 1912 1923 1928 1935 1942 1955 1967 1975 1980 1990 1998 2002 2012 2017 {
   replace cons = _b[_cons] if period==`p'
   replace b_fipr = _b[`p'.period#c.fipr] if period==`p'
   replace b_field = _b[`p'.period#c.field_composite] if period==`p'
   replace b_ifpop = _b[`p'.period#c.ifpop_pip] if period==`p'
   replace b_ofair = _b[`p'.period#c.ofair_pip] if period==`p'
}

 
 // save period-specific mean IF pop-up and OF airball rates
 
 // weightd means for batted ball variables; maybe average should be for period and not season in question?...
 * 1. Summarize IP within each period
egen total_IP = sum(IP), by(period)

 
* 2. Compute weighted mean of ifpop_pip
generate double weighted_ifpop_pip_product = ifpop_pip * IP
egen double total_weighted_ifpop_pip = sum(weighted_ifpop_pip_product), by(period)
generate double weighted_mean_ifpop_pip = total_weighted_ifpop_pip / total_IP




rename weighted_mean_ifpop_pip wvar_ifpop_pip





* 2. Compute weighted mean of ofair_pip
generate double weighted_ofair_pip_product = ofair_pip * IP
egen double total_weighted_ofair_pip = sum(weighted_ofair_pip_product), by(period)
generate double weighted_mean_ofair_pip = total_weighted_ofair_pip / total_IP



rename weighted_mean_ofair_pip wvar_ofair_pip

 
 
 
collapse wvar_ifpop wvar_ofair cons  b_fipr b_field b_ifpop b_ofair,by(period)


save [BIB_RBA parameters], replace