clear use Vars.dta, clear xtset id year foreach v in CE PGDP JA IS PS EC /// EI_EL IS_Ind1 ES_Fo RE REC CI PD CE90 { gen ln`v' = log(`v'+0.01) } gen lnFDI = cond(FDI > 0, ln(FDI+0.01), -ln(abs(FDI)+0.01)) global cv "lnPGDP lnJA lnIS lnPS lnFDI" global cvy "lnPGDP lnJA lnIS lnPS lnFDI i.year" **Table 1 Descriptive statistics. sum lnCE P_RE PT_RI PT_PS PT_EI PT_RD PT_VA PT_IE /// lnPGDP lnJA lnIS lnPS lnFDI /// RE lnEI_EL lnES_Fo lnPD lnIS_Ind1 lnEC lnREC /// CC RL BAtype BAtarget **Table 2 The carbon reduction effects of REPs. xtscc lnCE P_RE $cvy, fe lag(1) xtscc lnCE P_RE i.year, fe lag(1) xtscc lnCE PT_RI PT_PS PT_EI PT_RD PT_VA PT_IE $cvy, fe lag(1) **Table 3 Results of robustness analysis. xtscc lnCI P_RE $cvy /// , fe lag(1) xtscc lnEC P_RE $cvy /// , fe lag(1) xtscc lnCE P_RE lnPGDP lnJA lnIS_Ind1 lnPD lnFDI i.year /// , fe lag(1) xtscc lnCE P_RE $cvy lnEI_EL lnES_Fo i.year /// , fe lag(1) xtscc lnCE L.(P_RE $cv) i.year /// , fe lag(1) gen IV_PRE=lnCE90*L.P_RE xi:xtivreg2 lnCE $cvy /// (P_RE=L.P_RE IV_PRE), gmm2s fe robust winsor2 lnCE P_RE lnPGDP lnJA lnIS lnPS lnFDI, replace cuts(1 99) xtscc lnCE P_RE $cvy /// , fe lag(1) use Vars.dta, clear xtset id year foreach v in CE PGDP JA IS PS EC /// EI_EL IS_Ind1 ES_Fo RE REC CI PD CE90 { gen ln`v' = log(`v'+0.01) } gen lnFDI = cond(FDI > 0, ln(FDI+0.01), -ln(abs(FDI)+0.01)) global cv "lnPGDP lnJA lnIS lnPS lnFDI" global cvy "lnPGDP lnJA lnIS lnPS lnFDI i.year" **Table 4. Heterogeneity analyses of REPs on carbon emissions. center P_RE lnEI_EL CC RL BAtype BAtarget, replace gen P_RE_EI=c_P_RE*c_lnEI_EL gen P_RE_CC=c_P_RE*c_CC gen P_RE_RL=c_P_RE*c_RL gen P_RE_BAtype=c_P_RE*c_BAtype gen P_RE_BAtarget=c_P_RE*c_BAtarget xtscc lnCE P_RE lnEI_EL P_RE_EI lnPGDP lnJA lnIS lnPS lnFDI i.year /// , fe lag(1) xtscc lnCE P_RE CC P_RE_CC lnPGDP lnJA lnIS lnPS lnFDI i.year /// , fe lag(1) xtscc lnCE P_RE RL P_RE_RL lnPGDP lnJA lnIS lnPS lnFDI i.year /// , fe lag(1) xtscc lnCE P_RE BAtype P_RE_BAtype lnPGDP lnJA lnIS lnPS lnFDI i.year /// , fe lag(1) xtscc lnCE P_RE BAtarget P_RE_BAtarget lnPGDP lnJA lnIS lnPS lnFDI i.year /// , fe lag(1) **Table 5. Mediation effect results. sgmediation2 lnCE, mv(lnREC) iv(P_RE) /// cv(lnPGDP lnJA lnIS lnPS lnFDI i.id i.year) vce(bootstrap, reps(500)) sgmediation2 lnCE, mv(RE) iv(P_RE) /// cv(lnPGDP lnJA lnIS lnPS lnFDI i.id i.year) vce(bootstrap, reps(500)) reg lnCE P_RE lnREC lnPGDP lnJA lnIS lnPS lnFDI vif reg lnCE P_RE RE lnPGDP lnJA lnIS lnPS lnFDI vif **Table 6. Analyses of substitution effects. xtscc PFEC PREC i.year /// , fe lag(1) xtscc PFEC PREC lnPGDP lnJA lnIS lnPS lnFDI i.year /// , fe lag(1) xtscc PFoEP PREP i.year /// , fe lag(1) xtscc PFoEP PREP lnPGDP lnJA lnIS lnPS lnFDI i.year /// , fe lag(1) **Fig. 2 The lag effects of RES and REC on carbon emissions. xtscc lnCE P_RE L.RE $cvy, fe lag(1) xtscc lnCE P_RE L2.RE $cvy, fe lag(1) xtscc lnCE P_RE L3.RE $cvy, fe lag(1) xtscc lnCE P_RE L4.RE $cvy, fe lag(1) xtscc lnCE P_RE L5.RE $cvy, fe lag(1) xtscc lnCE P_RE L6.RE $cvy, fe lag(1) xtscc lnCE P_RE L7.RE $cvy, fe lag(1) xtscc lnCE P_RE L8.RE $cvy, fe lag(1) xtscc lnCE P_RE L9.RE $cvy, fe lag(1) xtscc lnCE P_RE L10.RE $cvy, fe lag(1) xtscc lnCE P_RE L11.RE $cvy, fe lag(1) xtscc lnCE P_RE L12.RE $cvy, fe lag(1) xtscc lnCE P_RE lnREC $cvy, fe lag(1) xtscc lnCE P_RE L.lnREC $cvy, fe lag(1) xtscc lnCE P_RE L2.lnREC $cvy, fe lag(1) xtscc lnCE P_RE L3.lnREC $cvy, fe lag(1) xtscc lnCE P_RE L4.lnREC $cvy, fe lag(1) xtscc lnCE P_RE L5.lnREC $cvy, fe lag(1) xtscc lnCE P_RE L6.lnREC $cvy, fe lag(1) xtscc lnCE P_RE L7.lnREC $cvy, fe lag(1) xtscc lnCE P_RE L8.lnREC $cvy, fe lag(1) xtscc lnCE P_RE L9.lnREC $cvy, fe lag(1) xtscc lnCE P_RE L10.lnREC $cvy, fe lag(1) xtscc lnCE P_RE L11.lnREC $cvy, fe lag(1) xtscc lnCE P_RE L12.lnREC $cvy, fe lag(1) **Fig.3 The heterogeneous effects of renewables on carbon emissions. *REP starting time xtscc lnCE P_RE RE lnPGDP lnJA lnIS lnPS lnFDI i.year if start2<5, fe lag(1) xtscc lnCE P_RE RE lnPGDP lnJA lnIS lnPS lnFDI i.year if start2>=5&start2<10, fe lag(1) xtscc lnCE P_RE RE lnPGDP lnJA lnIS lnPS lnFDI i.year if start2>=10&start2<15, fe lag(1) xtscc lnCE P_RE RE lnPGDP lnJA lnIS lnPS lnFDI i.year if start2>=15&start2<20, fe lag(1) xtscc lnCE P_RE RE lnPGDP lnJA lnIS lnPS lnFDI i.year if start2>=20, fe lag(1) *Geographical location xtscc lnCE P_RE RE $cvy if continent=="Asia", fe lag(1) xtscc lnCE P_RE RE $cvy if continent=="Europe", fe lag(1) xtscc lnCE P_RE RE $cvy if continent=="Africa", fe lag(1) xtscc lnCE P_RE RE $cvy if continent=="NorthAmerica", fe lag(1) xtscc lnCE P_RE RE $cvy if continent=="SouthAmerica", fe lag(1) xtscc lnCE P_RE RE $cvy if continent=="Oceania", fe lag(1) xtscc lnCE P_RE RE $cvy if continent=="Europe"&eu==0 , fe lag(1) *Organizational affiliation xtscc lnCE P_RE RE $cvy if eu==1, fe lag(1) xtscc lnCE P_RE RE $cvy if asean==1, fe lag(1) xtscc lnCE P_RE RE $cvy if africanunion==1, fe lag(1) xtscc lnCE P_RE RE $cvy if brics==1, fe lag(1) xtscc lnCE P_RE RE $cvy if opec==1, fe lag(1) xtscc lnCE P_RE RE $cvy if oecd==1, fe lag(1) *Economic level, industrialization level, population size, initial carbon emissions tabstat PGDP PS PIO2 , statistics(mean) tabstat CE if year==2000 , statistics(mean) /* stats | PGDP PS PIO2 ---------+------------------------------ mean | 13536.99 4.79e+07 29434.37 variable | mean -------------+---------- CE | 167563.3*/ xtscc lnCE P_RE RE $cvy if PGDP>13536.99& PGDP!= ., fe lag(1) xtscc lnCE P_RE RE $cvy if PGDP<13536.99, fe lag(1) xtscc lnCE P_RE RE $cvy if PIO2>29434.37& PIO2!= ., fe lag(1) xtscc lnCE P_RE RE $cvy if PIO2<29434.37, fe lag(1) xtscc lnCE P_RE RE $cvy if PS>4.79e+07& PS!= ., fe lag(1) xtscc lnCE P_RE RE $cvy if PS<4.79e+07, fe lag(1) gen CE2000 = (CE > 167563) if year == 2000 bysort id : egen min_CE2000 = min(CE2000) replace CE2000 = min_CE2000 xtscc lnCE P_RE RE $cvy if CE2000==1, fe lag(1) xtscc lnCE P_RE RE $cvy if CE2000==0, fe lag(1) **Table 7. The nonlinear relationship between RES and carbon reductions. gen RE2 =RE *RE xtscc lnCE P_RE RE RE2 lnPGDP lnJA lnIS lnPS lnFDI i.year /// , fe lag(1)