brycewang-stanford/Auto-Empirical-Research-Skills / skills/18-jusi-aalto-stata-accounting-research
stata-accounting-research
UnverifiedSTATA code pattern library for empirical archival accounting research. Provides tested syntax from 126 peer-reviewed JAR (Journal of Accounting Research) replication files (2017-2025). Use when the user asks procedural questions like "How do I implement [method]?" or "Show me code for [technique]" — including: entropy balancing, propensity score matching (PSM), difference-in-differences (DiD), regression discontinuity (RDD), instrumental variables (IV), event studies (CAR/BHAR), survival analysis, Fama-MacBeth regressions, bootstrap, quantile regression, reghdfe/xtreg/areg, clustering standard errors, fixed effects, esttab/outreg2 table formatting, winsorization, leads/lags. Users can specify their variables (e.g., treatment, outcomes, controls) and receive adapted syntax. NOTE: This skill provides code patterns from published papers, not research design advice.
references/JAR_55_mr.do
View on GitHub
*************************************************************************************************************
*DOES CORPORATE SOCIAL RESPONSIBILITY (CSR) CREATE SHAREHOLDER VALUE?
*EVIDENCE FROM THE INDIAN COMPANIES ACT 2013
*HARIOM MANCHIRAJU & SHIVA RAJGOPAL
*PARTS OF STATA CODE AND DESCRIPTION TO REPLICATE MAIN RESULTS - TABLE 3-6
****************************************************************************************************************
************************************************************************************************************
* STEP 1: GENERATE SAMPLE FOR THE RDD ANALYSIS
*The input dataset "acctdata" contains accountnig data for all firms with non-missing data on total assets,
*sales, net income, book value of equity, and market capitalization
*************************************************************************************************************
/*use acctdata
gen affected = 0
gen affected1 = 0
replace affected1 = 1 if pat >= 50
gen affected2 = 0
replace affected2 = 1 if bv >= 5000
gen affected3 = 0
replace affected3 = 1 if sale >= 10000
replace affected = 1 if affected1 == 1
replace affected = 1 if affected2 == 1
replace affected = 1 if affected3 == 1
gen bin1 = 0
replace bin1 = 1 if pat >= 25 & pat <= 75
gen bin2 = 0
replace bin2 = 1 if bv >= 2500 & bv <= 7500
gen bin3 = 0
replace bin3 = 1 if sales >= 5000 & sales <= 15000
gen rd_sample = 0
replace rd_sample = 1 if bin1 == 1
replace rd_sample = 1 if bin2 == 1
replace rd_sample = 1 if bin3 == 1
gen pat1 = (pat - 50)/50
gen bv1 = (bv - 5000)/5000
gen sales1 = (sales - 10000)/10000
gen m = 0
replace m = pat1 if affected1 == 1 & affected2 ==0 & affected3 == 0
replace m = bv1 if affected1 == 0 & affected2 ==1 & affected3 == 0
replace m = sales1 if affected1 == 0 & affected2 ==0 & affected3 == 1
replace m = min(pat1, bv1, sales1) if affected1 == 1 & affected2 ==1 & affected3 == 1
replace m = max(pat1, bv1, sales1) if affected1 == 0 & affected2 ==0 & affected3 == 0
replace m = min(pat1, bv1) if affected1 == 1 & affected2 ==1 & affected3 == 0
replace m = min(pat1, sales1) if affected1 == 1 & affected2 ==0 & affected3 ==1
replace m = min(bv1, sales1) if affected1 == 0 & affected2 ==1 & affected3 ==1
replace rd_sample = 0 if m > .5
keep if rd_sample == 1
save acctdata, replace
************************************************************************************************************
* STEP 2: GET STOCK RETURN DATA FOR THE RDD SAMPLE
*CAR is calculated using the market model around the event dates identified in the paper. The CAR data is then
*merged with accunting data using the unique key "CMIE COMPANY CODE" to create the final dataset "FINALDATA".
*************************************************************************************************************
**************************************************************************************************************
* STEP 3 - MAIN RESULTS
*The input dataset is "tables_3_6" contains CAR and accountnig data for the RDD sample
**************************************************************************************************************
**********************************************************************************************************
*GRAPHICAL ANALYSIS OF RDD - * FIGURE 3
**********************************************************************************************************
* This set of graphs will have scatter plot of the entire data in the background
*rd file needs to be installed */
cd "H:\Hariom\Research\Projects\8 Mandatory CSR\JAR\tables"
use tables_3_6
tab date, gen(event)
rd car3d m if event1 == 1, p(1) gr
rd car3d m if event2 == 1, p(1) gr
rd car3d m if event3 == 1, p(1) gr
rd car3d m if event4 == 1, p(1) gr
rd car3d m if event5 == 1, p(1) gr
rd car3d m if event6 == 1, p(1) gr
rd car3d m if event7 == 1, p(1) gr
rd car3d m if event8 == 1, p(1) gr
* coefficients and z-stats are tabulated in table 4a
* this set of graphs will have bin averages. The number of bins are also reported
*rdplot file needs to be installed
rdplot car3d m if event1 == 1, h(1) p(3)
rdplot car3d m if event2 == 1, h(1) p(3)
rdplot car3d m if event3 == 1, h(1) p(3)
rdplot car3d m if event4 == 1, h(1) p(3)
rdplot car3d m if event5 == 1, h(1) p(3)
rdplot car3d m if event6 == 1, h(1) p(3)
rdplot car3d m if event7 == 1, h(1) p(3)
rdplot car3d m if event8 == 1, h(1) p(3)
**********************************************************************************************************
* TABLE 3 - UNIVARIATE TESTS
**********************************************************************************************************
tabstat car3d if affected == 1, by(date) stat(median) col(stat)
tabstat car3d if affected == 0, by(date) stat(median) col(stat)
tabstat car3d if affected == 1 , by(date) stat(median) col(stat)
tabstat car3d if affected == 0 , by(date) stat(median) col(stat)
signrank car3d = 0 if event1 == 1 & affected == 0
signrank car3d = 0 if event1 == 1 & affected == 1
ranksum car3d if event1 == 1, by(affected)
signrank car3d = 0 if event2 == 1 & affected == 0
signrank car3d = 0 if event2 == 1 & affected == 1
ranksum car3d if event2 == 1 , by(affected)
signrank car3d = 0 if event3 == 1 & affected == 0
signrank car3d = 0 if event3 == 1 & affected == 1
ranksum car3d if event3 == 1 , by(affected)
signrank car3d = 0 if event4 == 1 & affected == 0
signrank car3d = 0 if event4 == 1 & affected == 1
ranksum car3d if event4 == 1 , by(affected)
signrank car3d = 0 if event5 == 1 & affected == 0
signrank car3d = 0 if event5 == 1 & affected == 1
ranksum car3d if event5 == 1 , by(affected)
signrank car3d = 0 if event6 == 1 & affected == 0
signrank car3d = 0 if event6 == 1 & affected == 1
ranksum car3d if event6 == 1 , by(affected)
signrank car3d = 0 if event7 == 1 & affected == 0
signrank car3d = 0 if event7 == 1 & affected == 1
ranksum car3d if event7 == 1 , by(affected)
signrank car3d = 0 if event8 == 1 & affected == 0
signrank car3d = 0 if event8 == 1 & affected == 1
ranksum car3d if event8 == 1 , by(affected)
**********************************************************************************************************
* TABLE 4 - DISCONTINUITY DESIGN - NONPARAMETRIC ESTIMATES
* RDROBUST ROUTINES of STATA NEED TO BE INSTALLED BEFORE RUNNING THIS PART OF THE CODE
**********************************************************************************************************
* Panel A - CAR as a linear function of M
rdrobust car3d m if event1 == 1, p(1) bwselect(ik)
rdrobust car3d m if event2 == 1, p(1) bwselect(ik)
rdrobust car3d m if event3 == 1, p(1) bwselect(ik)
rdrobust car3d m if event4 == 1, p(1) bwselect(ik)
rdrobust car3d m if event5 == 1, p(1) bwselect(ik)
rdrobust car3d m if event6 == 1, p(1) bwselect(ik)
rdrobust car3d m if event7 == 1, p(1) bwselect(ik)
rdrobust car3d m if event8 == 1, p(1) bwselect(ik)
* Panel B - CAR as a polynomial function of M
rdrobust car3d m if event1 == 1, h(1) p(3)
rdrobust car3d m if event2 == 1, h(1) p(3)
rdrobust car3d m if event3 == 1, h(1) p(3)
rdrobust car3d m if event4 == 1, h(1) p(3)
rdrobust car3d m if event5 == 1, h(1) p(3)
rdrobust car3d m if event6 == 1, h(1) p(3)
rdrobust car3d m if event7 == 1, h(1) p(3)
rdrobust car3d m if event8 == 1, h(1) p(3)
**********************************************************************************************************
* TABLE 5 - DISCONTINUITY DESIGN - MULTIPLE RUNNING VARIABLES - PARAMETRIC ESTIMATES WITH CONTROLS
* CLUSTER2 ROUTINE of STATA NEED TO BE INSTALLED BEFORE RUNNING THIS PART OF THE CODE
**********************************************************************************************************
local firm_controls size bm lev roa capex sgrowth cash bind big4 bg govt mnc xad political polluted
tab ffind, gen(ind)
gen affected_event1 = affected*event1
gen affected_event2 = affected*event2
gen affected_event3 = affected*event3
gen affected_event4 = affected*event4
gen affected_event5 = affected*event5
gen affected_event6 = affected*event6
gen affected_event7 = affected*event7
gen affected_event8 = affected*event8
gen affected_m = affected*m
gen m_p2 = m*m
gen m_p3 = m*m*m
cluster2 car3d affected_event1 affected_event2 affected_event3 affected_event4 affected_event5 affected_event6 affected_event7 affected_event8 affected_m m m_p2 m_p3 pat1 bv1 sales1 event2-event8, fcluster(coid) tcluster(date)
test affected_event1 + affected_event2 + affected_event3 + affected_event4 + affected_event5 + affected_event6 + affected_event7 + affected_event8 = 0
test affected_event1 + affected_event2 + affected_event3 + affected_event4 + affected_event5 + affected_event6 + affected_event7 + affected_event8 = -.014
outreg2 using t5.xls, stats( coef tstat) aster paren dec(3) sideway
cluster2 car3d affected_event1 affected_event2 affected_event3 affected_event4 affected_event5 affected_event6 affected_event7 affected_event8 affected_m m m_p2 m_p3 pat1 bv1 sales1 `firm_controls' ind1-ind38 event2-event8, fcluster(coid) tcluster(date)
test affected_event1 + affected_event2 + affected_event3 + affected_event4 + affected_event5 + affected_event6 + affected_event7 + affected_event8 = 0
test affected_event1 + affected_event2 + affected_event3 + affected_event4 + affected_event5 + affected_event6 + affected_event7 + affected_event8 = -.014
outreg2 using t5.xls, stats( coef tstat) aster paren dec(3) sideway
areg car3d affected_event1 affected_event2 affected_event3 affected_event4 affected_event5 affected_event6 affected_event7 affected_event8 affected_m m m_p2 m_p3 pat1 bv1 sales1 `firm_controls' event2-event8, cluster(coid) absorb(coid)
test affected_event1 + affected_event2 + affected_event3 + affected_event4 + affected_event5 + affected_event6 + affected_event7 + affected_event8 = 0
test affected_event1 + affected_event2 + affected_event3 + affected_event4 + affected_event5 + affected_event6 + affected_event7 + affected_event8 = -.014
outreg2 using t5.xls, stats( coef tstat) aster paren dec(3) sideway
**********************************************************************************************************
* TABLE 6 - CROSS SECTIONAL ANALYSIS - RDD WITH PARAMETRIC ESTIMATES WITH CONTROLS
* CLUSTER2 ROUTINE of STATA NEED TO BE INSTALLED BEFORE RUNNING THIS PART OF THE CODE
**********************************************************************************************************
* VARIATION BASED ON THE BUSINESS GROUP AFFILIATION
gen affected_event1_political = affected*event1*political
gen affected_event2_political = affected*event2*political
gen affected_event3_political = affected*event3*political
gen affected_event4_political = affected*event4*political
gen affected_event5_political = affected*event5*political
gen affected_event6_political = affected*event6*political
gen affected_event7_political = affected*event7*political
gen affected_event8_political = affected*event8*political
cluster2 car3d affected_event1 affected_event2 affected_event3 affected_event4 affected_event5 affected_event6 affected_event7 affected_event8 affected_event1_political affected_event2_political affected_event3_political affected_event4_political affected_event5_political affected_event6_political affected_event7_political affected_event8_political affected_m m m_p2 m_p3 pat1 bv1 sales1 event1-event8 `firm_controls' ind1-ind38, fcluster(coid) tcluster(date)
test affected_event1_political + affected_event2_political + affected_event3_political + affected_event4_political + affected_event5_political + affected_event6_political + affected_event7_political + affected_event8_political = 0
outreg2 using t6.xls, stats( coef tstat) aster paren dec(3) sideway
* VARIATION BASED ON HIGH ADVERTISEMENT SPENDERS
gen affected_event1_ad = affected*event1*ad
gen affected_event2_ad = affected*event2*ad
gen affected_event3_ad = affected*event3*ad
gen affected_event4_ad = affected*event4*ad
gen affected_event5_ad = affected*event5*ad
gen affected_event6_ad = affected*event6*ad
gen affected_event7_ad = affected*event7*ad
gen affected_event8_ad = affected*event8*ad
cluster2 car3d affected_event1 affected_event2 affected_event3 affected_event4 affected_event5 affected_event6 affected_event7 affected_event8 affected_event1_ad affected_event2_ad affected_event3_ad affected_event4_ad affected_event5_ad affected_event6_ad affected_event7_ad affected_event8_ad affected_m m m_p2 m_p3 pat1 bv1 sales1 event1-event8 `firm_controls' ind1-ind38, fcluster(coid) tcluster(date)
test affected_event1_ad + affected_event2_ad + affected_event3_ad + affected_event4_ad + affected_event5_ad + affected_event6_ad + affected_event7_ad + affected_event8_ad = 0
outreg2 using t6.xls, stats( coef tstat) aster paren dec(3) sideway
gen affected_event1_polluted = affected*event1*polluted
gen affected_event2_polluted = affected*event2*polluted
gen affected_event3_polluted = affected*event3*polluted
gen affected_event4_polluted = affected*event4*polluted
gen affected_event5_polluted = affected*event5*polluted
gen affected_event6_polluted = affected*event6*polluted
gen affected_event7_polluted = affected*event7*polluted
gen affected_event8_polluted = affected*event8*polluted
cluster2 car3d affected_event1 affected_event2 affected_event3 affected_event4 affected_event5 affected_event6 affected_event7 affected_event8 affected_event1_polluted affected_event2_polluted affected_event3_polluted affected_event4_polluted affected_event5_polluted affected_event6_polluted affected_event7_polluted affected_event8_polluted affected_m m m_p2 m_p3 pat1 bv1 sales1 event1-event8 `firm_controls' ind1-ind38, fcluster(coid) tcluster(date)
test affected_event1_polluted + affected_event2_polluted + affected_event3_polluted + affected_event4_polluted + affected_event5_polluted + affected_event6_polluted + affected_event7_polluted + affected_event8_polluted = 0
outreg2 using t6.xls, stats( coef tstat) aster paren dec(3) sideway
***************************************************************************************************************************************************************
*************************************************************************************************************
*DOES CORPORATE SOCIAL RESPONSIBILITY (CSR) CREATE SHAREHOLDER VALUE?
*EVIDENCE FROM THE INDIAN COMPANIES ACT 2013
*HARIOM MANCHIRAJU & SHIVA RAJGOPAL
*PARTS OF STATA CODE AND DESCRIPTION TO REPLICATE MAIN RESULTS - TABLE 3-6
****************************************************************************************************************
************************************************************************************************************
* STEP 1: GENERATE SAMPLE FOR THE RDD ANALYSIS
*The input dataset "acctdata" contains accountnig data for all firms with non-missing data on total assets,
*sales, net income, book value of equity, and market capitalization
*************************************************************************************************************
/*use acctdata
gen affected = 0
gen affected1 = 0
replace affected1 = 1 if pat >= 50
gen affected2 = 0
replace affected2 = 1 if bv >= 5000
gen affected3 = 0
replace affected3 = 1 if sale >= 10000
replace affected = 1 if affected1 == 1
replace affected = 1 if affected2 == 1
replace affected = 1 if affected3 == 1
gen bin1 = 0
replace bin1 = 1 if pat >= 25 & pat <= 75
gen bin2 = 0
replace bin2 = 1 if bv >= 2500 & bv <= 7500
gen bin3 = 0
replace bin3 = 1 if sales >= 5000 & sales <= 15000
gen rd_sample = 0
replace rd_sample = 1 if bin1 == 1
replace rd_sample = 1 if bin2 == 1
replace rd_sample = 1 if bin3 == 1
gen pat1 = (pat - 50)/50
gen bv1 = (bv - 5000)/5000
gen sales1 = (sales - 10000)/10000
gen m = 0
replace m = pat1 if affected1 == 1 & affected2 ==0 & affected3 == 0
replace m = bv1 if affected1 == 0 & affected2 ==1 & affected3 == 0
replace m = sales1 if affected1 == 0 & affected2 ==0 & affected3 == 1
replace m = min(pat1, bv1, sales1) if affected1 == 1 & affected2 ==1 & affected3 == 1
replace m = max(pat1, bv1, sales1) if affected1 == 0 & affected2 ==0 & affected3 == 0
replace m = min(pat1, bv1) if affected1 == 1 & affected2 ==1 & affected3 == 0
replace m = min(pat1, sales1) if affected1 == 1 & affected2 ==0 & affected3 ==1
replace m = min(bv1, sales1) if affected1 == 0 & affected2 ==1 & affected3 ==1
replace rd_sample = 0 if m > .5
keep if rd_sample == 1
save acctdata, replace
************************************************************************************************************
* STEP 2: GET STOCK RETURN DATA FOR THE RDD SAMPLE
*CAR is calculated using the market model around the event dates identified in the paper. The CAR data is then
*merged with accunting data using the unique key "CMIE COMPANY CODE" to create the final dataset "FINALDATA".
*************************************************************************************************************
**************************************************************************************************************
* STEP 3 - MAIN RESULTS
*The input dataset is "tables_3_6" contains CAR and accountnig data for the RDD sample
**************************************************************************************************************
**********************************************************************************************************
*GRAPHICAL ANALYSIS OF RDD - * FIGURE 3
**********************************************************************************************************
* This set of graphs will have scatter plot of the entire data in the background
*rd file needs to be installed */
cd "H:\Hariom\Research\Projects\8 Mandatory CSR\JAR\tables"
use tables_3_6
tab date, gen(event)
rd car3d m if event1 == 1, p(1) gr
rd car3d m if event2 == 1, p(1) gr
rd car3d m if event3 == 1, p(1) gr
rd car3d m if event4 == 1, p(1) gr
rd car3d m if event5 == 1, p(1) gr
rd car3d m if event6 == 1, p(1) gr
rd car3d m if event7 == 1, p(1) gr
rd car3d m if event8 == 1, p(1) gr
* coefficients and z-stats are tabulated in table 4a
* this set of graphs will have bin averages. The number of bins are also reported
*rdplot file needs to be installed
rdplot car3d m if event1 == 1, h(1) p(3)
rdplot car3d m if event2 == 1, h(1) p(3)
rdplot car3d m if event3 == 1, h(1) p(3)
rdplot car3d m if event4 == 1, h(1) p(3)
rdplot car3d m if event5 == 1, h(1) p(3)
rdplot car3d m if event6 == 1, h(1) p(3)
rdplot car3d m if event7 == 1, h(1) p(3)
rdplot car3d m if event8 == 1, h(1) p(3)
**********************************************************************************************************
* TABLE 3 - UNIVARIATE TESTS
**********************************************************************************************************
tabstat car3d if affected == 1, by(date) stat(median) col(stat)
tabstat car3d if affected == 0, by(date) stat(median) col(stat)
tabstat car3d if affected == 1 , by(date) stat(median) col(stat)
tabstat car3d if affected == 0 , by(date) stat(median) col(stat)
signrank car3d = 0 if event1 == 1 & affected == 0
signrank car3d = 0 if event1 == 1 & affected == 1
ranksum car3d if event1 == 1, by(affected)
signrank car3d = 0 if event2 == 1 & affected == 0
signrank car3d = 0 if event2 == 1 & affected == 1
ranksum car3d if event2 == 1 , by(affected)
signrank car3d = 0 if event3 == 1 & affected == 0
signrank car3d = 0 if event3 == 1 & affected == 1
ranksum car3d if event3 == 1 , by(affected)
signrank car3d = 0 if event4 == 1 & affected == 0
signrank car3d = 0 if event4 == 1 & affected == 1
ranksum car3d if event4 == 1 , by(affected)
signrank car3d = 0 if event5 == 1 & affected == 0
signrank car3d = 0 if event5 == 1 & affected == 1
ranksum car3d if event5 == 1 , by(affected)
signrank car3d = 0 if event6 == 1 & affected == 0
signrank car3d = 0 if event6 == 1 & affected == 1
ranksum car3d if event6 == 1 , by(affected)
signrank car3d = 0 if event7 == 1 & affected == 0
signrank car3d = 0 if event7 == 1 & affected == 1
ranksum car3d if event7 == 1 , by(affected)
signrank car3d = 0 if event8 == 1 & affected == 0
signrank car3d = 0 if event8 == 1 & affected == 1
ranksum car3d if event8 == 1 , by(affected)
**********************************************************************************************************
* TABLE 4 - DISCONTINUITY DESIGN - NONPARAMETRIC ESTIMATES
* RDROBUST ROUTINES of STATA NEED TO BE INSTALLED BEFORE RUNNING THIS PART OF THE CODE
**********************************************************************************************************
* Panel A - CAR as a linear function of M
rdrobust car3d m if event1 == 1, p(1) bwselect(ik)
rdrobust car3d m if event2 == 1, p(1) bwselect(ik)
rdrobust car3d m if event3 == 1, p(1) bwselect(ik)
rdrobust car3d m if event4 == 1, p(1) bwselect(ik)
rdrobust car3d m if event5 == 1, p(1) bwselect(ik)
rdrobust car3d m if event6 == 1, p(1) bwselect(ik)
rdrobust car3d m if event7 == 1, p(1) bwselect(ik)
rdrobust car3d m if event8 == 1, p(1) bwselect(ik)
* Panel B - CAR as a polynomial function of M
rdrobust car3d m if event1 == 1, h(1) p(3)
rdrobust car3d m if event2 == 1, h(1) p(3)
rdrobust car3d m if event3 == 1, h(1) p(3)
rdrobust car3d m if event4 == 1, h(1) p(3)
rdrobust car3d m if event5 == 1, h(1) p(3)
rdrobust car3d m if event6 == 1, h(1) p(3)
rdrobust car3d m if event7 == 1, h(1) p(3)
rdrobust car3d m if event8 == 1, h(1) p(3)
**********************************************************************************************************
* TABLE 5 - DISCONTINUITY DESIGN - MULTIPLE RUNNING VARIABLES - PARAMETRIC ESTIMATES WITH CONTROLS
* CLUSTER2 ROUTINE of STATA NEED TO BE INSTALLED BEFORE RUNNING THIS PART OF THE CODE
**********************************************************************************************************
local firm_controls size bm lev roa capex sgrowth cash bind big4 bg govt mnc xad political polluted
tab ffind, gen(ind)
gen affected_event1 = affected*event1
gen affected_event2 = affected*event2
gen affected_event3 = affected*event3
gen affected_event4 = affected*event4
gen affected_event5 = affected*event5
gen affected_event6 = affected*event6
gen affected_event7 = affected*event7
gen affected_event8 = affected*event8
gen affected_m = affected*m
gen m_p2 = m*m
gen m_p3 = m*m*m
cluster2 car3d affected_event1 affected_event2 affected_event3 affected_event4 affected_event5 affected_event6 affected_event7 affected_event8 affected_m m m_p2 m_p3 pat1 bv1 sales1 event2-event8, fcluster(coid) tcluster(date)
test affected_event1 + affected_event2 + affected_event3 + affected_event4 + affected_event5 + affected_event6 + affected_event7 + affected_event8 = 0
test affected_event1 + affected_event2 + affected_event3 + affected_event4 + affected_event5 + affected_event6 + affected_event7 + affected_event8 = -.014
outreg2 using t5.xls, stats( coef tstat) aster paren dec(3) sideway
cluster2 car3d affected_event1 affected_event2 affected_event3 affected_event4 affected_event5 affected_event6 affected_event7 affected_event8 affected_m m m_p2 m_p3 pat1 bv1 sales1 `firm_controls' ind1-ind38 event2-event8, fcluster(coid) tcluster(date)
test affected_event1 + affected_event2 + affected_event3 + affected_event4 + affected_event5 + affected_event6 + affected_event7 + affected_event8 = 0
test affected_event1 + affected_event2 + affected_event3 + affected_event4 + affected_event5 + affected_event6 + affected_event7 + affected_event8 = -.014
outreg2 using t5.xls, stats( coef tstat) aster paren dec(3) sideway
areg car3d affected_event1 affected_event2 affected_event3 affected_event4 affected_event5 affected_event6 affected_event7 affected_event8 affected_m m m_p2 m_p3 pat1 bv1 sales1 `firm_controls' event2-event8, cluster(coid) absorb(coid)
test affected_event1 + affected_event2 + affected_event3 + affected_event4 + affected_event5 + affected_event6 + affected_event7 + affected_event8 = 0
test affected_event1 + affected_event2 + affected_event3 + affected_event4 + affected_event5 + affected_event6 + affected_event7 + affected_event8 = -.014
outreg2 using t5.xls, stats( coef tstat) aster paren dec(3) sideway
**********************************************************************************************************
* TABLE 6 - CROSS SECTIONAL ANALYSIS - RDD WITH PARAMETRIC ESTIMATES WITH CONTROLS
* CLUSTER2 ROUTINE of STATA NEED TO BE INSTALLED BEFORE RUNNING THIS PART OF THE CODE
**********************************************************************************************************
* VARIATION BASED ON THE BUSINESS GROUP AFFILIATION
gen affected_event1_political = affected*event1*political
gen affected_event2_political = affected*event2*political
gen affected_event3_political = affected*event3*political
gen affected_event4_political = affected*event4*political
gen affected_event5_political = affected*event5*political
gen affected_event6_political = affected*event6*political
gen affected_event7_political = affected*event7*political
gen affected_event8_political = affected*event8*political
cluster2 car3d affected_event1 affected_event2 affected_event3 affected_event4 affected_event5 affected_event6 affected_event7 affected_event8 affected_event1_political affected_event2_political affected_event3_political affected_event4_political affected_event5_political affected_event6_political affected_event7_political affected_event8_political affected_m m m_p2 m_p3 pat1 bv1 sales1 event1-event8 `firm_controls' ind1-ind38, fcluster(coid) tcluster(date)
test affected_event1_political + affected_event2_political + affected_event3_political + affected_event4_political + affected_event5_political + affected_event6_political + affected_event7_political + affected_event8_political = 0
outreg2 using t6.xls, stats( coef tstat) aster paren dec(3) sideway
* VARIATION BASED ON HIGH ADVERTISEMENT SPENDERS
gen affected_event1_ad = affected*event1*ad
gen affected_event2_ad = affected*event2*ad
gen affected_event3_ad = affected*event3*ad
gen affected_event4_ad = affected*event4*ad
gen affected_event5_ad = affected*event5*ad
gen affected_event6_ad = affected*event6*ad
gen affected_event7_ad = affected*event7*ad
gen affected_event8_ad = affected*event8*ad
cluster2 car3d affected_event1 affected_event2 affected_event3 affected_event4 affected_event5 affected_event6 affected_event7 affected_event8 affected_event1_ad affected_event2_ad affected_event3_ad affected_event4_ad affected_event5_ad affected_event6_ad affected_event7_ad affected_event8_ad affected_m m m_p2 m_p3 pat1 bv1 sales1 event1-event8 `firm_controls' ind1-ind38, fcluster(coid) tcluster(date)
test affected_event1_ad + affected_event2_ad + affected_event3_ad + affected_event4_ad + affected_event5_ad + affected_event6_ad + affected_event7_ad + affected_event8_ad = 0
outreg2 using t6.xls, stats( coef tstat) aster paren dec(3) sideway
gen affected_event1_polluted = affected*event1*polluted
gen affected_event2_polluted = affected*event2*polluted
gen affected_event3_polluted = affected*event3*polluted
gen affected_event4_polluted = affected*event4*polluted
gen affected_event5_polluted = affected*event5*polluted
gen affected_event6_polluted = affected*event6*polluted
gen affected_event7_polluted = affected*event7*polluted
gen affected_event8_polluted = affected*event8*polluted
cluster2 car3d affected_event1 affected_event2 affected_event3 affected_event4 affected_event5 affected_event6 affected_event7 affected_event8 affected_event1_polluted affected_event2_polluted affected_event3_polluted affected_event4_polluted affected_event5_polluted affected_event6_polluted affected_event7_polluted affected_event8_polluted affected_m m m_p2 m_p3 pat1 bv1 sales1 event1-event8 `firm_controls' ind1-ind38, fcluster(coid) tcluster(date)
test affected_event1_polluted + affected_event2_polluted + affected_event3_polluted + affected_event4_polluted + affected_event5_polluted + affected_event6_polluted + affected_event7_polluted + affected_event8_polluted = 0
outreg2 using t6.xls, stats( coef tstat) aster paren dec(3) sideway
***************************************************************************************************************************************************************
*************************************************************************************************************
*DOES CORPORATE SOCIAL RESPONSIBILITY (CSR) CREATE SHAREHOLDER VALUE?
*EVIDENCE FROM THE INDIAN COMPANIES ACT 2013
*HARIOM MANCHIRAJU & SHIVA RAJGOPAL
*PARTS OF STATA CODE AND DESCRIPTION TO REPLICATE MAIN RESULTS - TABLE 3-6
****************************************************************************************************************
************************************************************************************************************
* STEP 1: GENERATE SAMPLE FOR THE RDD ANALYSIS
*The input dataset "acctdata" contains accountnig data for all firms with non-missing data on total assets,
*sales, net income, book value of equity, and market capitalization
*************************************************************************************************************
/*use acctdata
gen affected = 0
gen affected1 = 0
replace affected1 = 1 if pat >= 50
gen affected2 = 0
replace affected2 = 1 if bv >= 5000
gen affected3 = 0
replace affected3 = 1 if sale >= 10000
replace affected = 1 if affected1 == 1
replace affected = 1 if affected2 == 1
replace affected = 1 if affected3 == 1
gen bin1 = 0
replace bin1 = 1 if pat >= 25 & pat <= 75
gen bin2 = 0
replace bin2 = 1 if bv >= 2500 & bv <= 7500
gen bin3 = 0
replace bin3 = 1 if sales >= 5000 & sales <= 15000
gen rd_sample = 0
replace rd_sample = 1 if bin1 == 1
replace rd_sample = 1 if bin2 == 1
replace rd_sample = 1 if bin3 == 1
gen pat1 = (pat - 50)/50
gen bv1 = (bv - 5000)/5000
gen sales1 = (sales - 10000)/10000
gen m = 0
replace m = pat1 if affected1 == 1 & affected2 ==0 & affected3 == 0
replace m = bv1 if affected1 == 0 & affected2 ==1 & affected3 == 0
replace m = sales1 if affected1 == 0 & affected2 ==0 & affected3 == 1
replace m = min(pat1, bv1, sales1) if affected1 == 1 & affected2 ==1 & affected3 == 1
replace m = max(pat1, bv1, sales1) if affected1 == 0 & affected2 ==0 & affected3 == 0
replace m = min(pat1, bv1) if affected1 == 1 & affected2 ==1 & affected3 == 0
replace m = min(pat1, sales1) if affected1 == 1 & affected2 ==0 & affected3 ==1
replace m = min(bv1, sales1) if affected1 == 0 & affected2 ==1 & affected3 ==1
replace rd_sample = 0 if m > .5
keep if rd_sample == 1
save acctdata, replace
************************************************************************************************************
* STEP 2: GET STOCK RETURN DATA FOR THE RDD SAMPLE
*CAR is calculated using the market model around the event dates identified in the paper. The CAR data is then
*merged with accunting data using the unique key "CMIE COMPANY CODE" to create the final dataset "FINALDATA".
*************************************************************************************************************
**************************************************************************************************************
* STEP 3 - MAIN RESULTS
*The input dataset is "tables_3_6" contains CAR and accountnig data for the RDD sample
**************************************************************************************************************
**********************************************************************************************************
*GRAPHICAL ANALYSIS OF RDD - * FIGURE 3
**********************************************************************************************************
* This set of graphs will have scatter plot of the entire data in the background
*rd file needs to be installed */
cd "H:\Hariom\Research\Projects\8 Mandatory CSR\JAR\tables"
use tables_3_6
tab date, gen(event)
rd car3d m if event1 == 1, p(1) gr
rd car3d m if event2 == 1, p(1) gr
rd car3d m if event3 == 1, p(1) gr
rd car3d m if event4 == 1, p(1) gr
rd car3d m if event5 == 1, p(1) gr
rd car3d m if event6 == 1, p(1) gr
rd car3d m if event7 == 1, p(1) gr
rd car3d m if event8 == 1, p(1) gr
* coefficients and z-stats are tabulated in table 4a
* this set of graphs will have bin averages. The number of bins are also reported
*rdplot file needs to be installed
rdplot car3d m if event1 == 1, h(1) p(3)
rdplot car3d m if event2 == 1, h(1) p(3)
rdplot car3d m if event3 == 1, h(1) p(3)
rdplot car3d m if event4 == 1, h(1) p(3)
rdplot car3d m if event5 == 1, h(1) p(3)
rdplot car3d m if event6 == 1, h(1) p(3)
rdplot car3d m if event7 == 1, h(1) p(3)
rdplot car3d m if event8 == 1, h(1) p(3)
**********************************************************************************************************
* TABLE 3 - UNIVARIATE TESTS
**********************************************************************************************************
tabstat car3d if affected == 1, by(date) stat(median) col(stat)
tabstat car3d if affected == 0, by(date) stat(median) col(stat)
tabstat car3d if affected == 1 , by(date) stat(median) col(stat)
tabstat car3d if affected == 0 , by(date) stat(median) col(stat)
signrank car3d = 0 if event1 == 1 & affected == 0
signrank car3d = 0 if event1 == 1 & affected == 1
ranksum car3d if event1 == 1, by(affected)
signrank car3d = 0 if event2 == 1 & affected == 0
signrank car3d = 0 if event2 == 1 & affected == 1
ranksum car3d if event2 == 1 , by(affected)
signrank car3d = 0 if event3 == 1 & affected == 0
signrank car3d = 0 if event3 == 1 & affected == 1
ranksum car3d if event3 == 1 , by(affected)
signrank car3d = 0 if event4 == 1 & affected == 0
signrank car3d = 0 if event4 == 1 & affected == 1
ranksum car3d if event4 == 1 , by(affected)
signrank car3d = 0 if event5 == 1 & affected == 0
signrank car3d = 0 if event5 == 1 & affected == 1
ranksum car3d if event5 == 1 , by(affected)
signrank car3d = 0 if event6 == 1 & affected == 0
signrank car3d = 0 if event6 == 1 & affected == 1
ranksum car3d if event6 == 1 , by(affected)
signrank car3d = 0 if event7 == 1 & affected == 0
signrank car3d = 0 if event7 == 1 & affected == 1
ranksum car3d if event7 == 1 , by(affected)
signrank car3d = 0 if event8 == 1 & affected == 0
signrank car3d = 0 if event8 == 1 & affected == 1
ranksum car3d if event8 == 1 , by(affected)
**********************************************************************************************************
* TABLE 4 - DISCONTINUITY DESIGN - NONPARAMETRIC ESTIMATES
* RDROBUST ROUTINES of STATA NEED TO BE INSTALLED BEFORE RUNNING THIS PART OF THE CODE
**********************************************************************************************************
* Panel A - CAR as a linear function of M
rdrobust car3d m if event1 == 1, p(1) bwselect(ik)
rdrobust car3d m if event2 == 1, p(1) bwselect(ik)
rdrobust car3d m if event3 == 1, p(1) bwselect(ik)
rdrobust car3d m if event4 == 1, p(1) bwselect(ik)
rdrobust car3d m if event5 == 1, p(1) bwselect(ik)
rdrobust car3d m if event6 == 1, p(1) bwselect(ik)
rdrobust car3d m if event7 == 1, p(1) bwselect(ik)
rdrobust car3d m if event8 == 1, p(1) bwselect(ik)
* Panel B - CAR as a polynomial function of M
rdrobust car3d m if event1 == 1, h(1) p(3)
rdrobust car3d m if event2 == 1, h(1) p(3)
rdrobust car3d m if event3 == 1, h(1) p(3)
rdrobust car3d m if event4 == 1, h(1) p(3)
rdrobust car3d m if event5 == 1, h(1) p(3)
rdrobust car3d m if event6 == 1, h(1) p(3)
rdrobust car3d m if event7 == 1, h(1) p(3)
rdrobust car3d m if event8 == 1, h(1) p(3)
**********************************************************************************************************
* TABLE 5 - DISCONTINUITY DESIGN - MULTIPLE RUNNING VARIABLES - PARAMETRIC ESTIMATES WITH CONTROLS
* CLUSTER2 ROUTINE of STATA NEED TO BE INSTALLED BEFORE RUNNING THIS PART OF THE CODE
**********************************************************************************************************
local firm_controls size bm lev roa capex sgrowth cash bind big4 bg govt mnc xad political polluted
tab ffind, gen(ind)
gen affected_event1 = affected*event1
gen affected_event2 = affected*event2
gen affected_event3 = affected*event3
gen affected_event4 = affected*event4
gen affected_event5 = affected*event5
gen affected_event6 = affected*event6
gen affected_event7 = affected*event7
gen affected_event8 = affected*event8
gen affected_m = affected*m
gen m_p2 = m*m
gen m_p3 = m*m*m
cluster2 car3d affected_event1 affected_event2 affected_event3 affected_event4 affected_event5 affected_event6 affected_event7 affected_event8 affected_m m m_p2 m_p3 pat1 bv1 sales1 event2-event8, fcluster(coid) tcluster(date)
test affected_event1 + affected_event2 + affected_event3 + affected_event4 + affected_event5 + affected_event6 + affected_event7 + affected_event8 = 0
test affected_event1 + affected_event2 + affected_event3 + affected_event4 + affected_event5 + affected_event6 + affected_event7 + affected_event8 = -.014
outreg2 using t5.xls, stats( coef tstat) aster paren dec(3) sideway
cluster2 car3d affected_event1 affected_event2 affected_event3 affected_event4 affected_event5 affected_event6 affected_event7 affected_event8 affected_m m m_p2 m_p3 pat1 bv1 sales1 `firm_controls' ind1-ind38 event2-event8, fcluster(coid) tcluster(date)
test affected_event1 + affected_event2 + affected_event3 + affected_event4 + affected_event5 + affected_event6 + affected_event7 + affected_event8 = 0
test affected_event1 + affected_event2 + affected_event3 + affected_event4 + affected_event5 + affected_event6 + affected_event7 + affected_event8 = -.014
outreg2 using t5.xls, stats( coef tstat) aster paren dec(3) sideway
areg car3d affected_event1 affected_event2 affected_event3 affected_event4 affected_event5 affected_event6 affected_event7 affected_event8 affected_m m m_p2 m_p3 pat1 bv1 sales1 `firm_controls' event2-event8, cluster(coid) absorb(coid)
test affected_event1 + affected_event2 + affected_event3 + affected_event4 + affected_event5 + affected_event6 + affected_event7 + affected_event8 = 0
test affected_event1 + affected_event2 + affected_event3 + affected_event4 + affected_event5 + affected_event6 + affected_event7 + affected_event8 = -.014
outreg2 using t5.xls, stats( coef tstat) aster paren dec(3) sideway
**********************************************************************************************************
* TABLE 6 - CROSS SECTIONAL ANALYSIS - RDD WITH PARAMETRIC ESTIMATES WITH CONTROLS
* CLUSTER2 ROUTINE of STATA NEED TO BE INSTALLED BEFORE RUNNING THIS PART OF THE CODE
**********************************************************************************************************
* VARIATION BASED ON THE BUSINESS GROUP AFFILIATION
gen affected_event1_political = affected*event1*political
gen affected_event2_political = affected*event2*political
gen affected_event3_political = affected*event3*political
gen affected_event4_political = affected*event4*political
gen affected_event5_political = affected*event5*political
gen affected_event6_political = affected*event6*political
gen affected_event7_political = affected*event7*political
gen affected_event8_political = affected*event8*political
cluster2 car3d affected_event1 affected_event2 affected_event3 affected_event4 affected_event5 affected_event6 affected_event7 affected_event8 affected_event1_political affected_event2_political affected_event3_political affected_event4_political affected_event5_political affected_event6_political affected_event7_political affected_event8_political affected_m m m_p2 m_p3 pat1 bv1 sales1 event1-event8 `firm_controls' ind1-ind38, fcluster(coid) tcluster(date)
test affected_event1_political + affected_event2_political + affected_event3_political + affected_event4_political + affected_event5_political + affected_event6_political + affected_event7_political + affected_event8_political = 0
outreg2 using t6.xls, stats( coef tstat) aster paren dec(3) sideway
* VARIATION BASED ON HIGH ADVERTISEMENT SPENDERS
gen affected_event1_ad = affected*event1*ad
gen affected_event2_ad = affected*event2*ad
gen affected_event3_ad = affected*event3*ad
gen affected_event4_ad = affected*event4*ad
gen affected_event5_ad = affected*event5*ad
gen affected_event6_ad = affected*event6*ad
gen affected_event7_ad = affected*event7*ad
gen affected_event8_ad = affected*event8*ad
cluster2 car3d affected_event1 affected_event2 affected_event3 affected_event4 affected_event5 affected_event6 affected_event7 affected_event8 affected_event1_ad affected_event2_ad affected_event3_ad affected_event4_ad affected_event5_ad affected_event6_ad affected_event7_ad affected_event8_ad affected_m m m_p2 m_p3 pat1 bv1 sales1 event1-event8 `firm_controls' ind1-ind38, fcluster(coid) tcluster(date)
test affected_event1_ad + affected_event2_ad + affected_event3_ad + affected_event4_ad + affected_event5_ad + affected_event6_ad + affected_event7_ad + affected_event8_ad = 0
outreg2 using t6.xls, stats( coef tstat) aster paren dec(3) sideway
gen affected_event1_polluted = affected*event1*polluted
gen affected_event2_polluted = affected*event2*polluted
gen affected_event3_polluted = affected*event3*polluted
gen affected_event4_polluted = affected*event4*polluted
gen affected_event5_polluted = affected*event5*polluted
gen affected_event6_polluted = affected*event6*polluted
gen affected_event7_polluted = affected*event7*polluted
gen affected_event8_polluted = affected*event8*polluted
cluster2 car3d affected_event1 affected_event2 affected_event3 affected_event4 affected_event5 affected_event6 affected_event7 affected_event8 affected_event1_polluted affected_event2_polluted affected_event3_polluted affected_event4_polluted affected_event5_polluted affected_event6_polluted affected_event7_polluted affected_event8_polluted affected_m m m_p2 m_p3 pat1 bv1 sales1 event1-event8 `firm_controls' ind1-ind38, fcluster(coid) tcluster(date)
test affected_event1_polluted + affected_event2_polluted + affected_event3_polluted + affected_event4_polluted + affected_event5_polluted + affected_event6_polluted + affected_event7_polluted + affected_event8_polluted = 0
outreg2 using t6.xls, stats( coef tstat) aster paren dec(3) sideway
***************************************************************************************************************************************************************
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| brainstormingobra/superpowersYou MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation). | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| brand-guidelinesanthropics/skillsApplies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| frontend-designanthropics/skillsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| mcp-builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK). | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |