------------------------------------------------------------------------------- name: log: /Users/carlosmendez/Documents/GitHub/starter-academic-v501/content > /post/stata_cate2/analysis.log log type: text opened on: 6 May 2026, 17:29:21 . . . *================================================================ . * Section 0: Stata 19 version gate . *================================================================ . * . * The -cate- command is brand-new in Stata 19. There is NO . * equivalent in Stata 18 or earlier. We refuse to run on older . * Stata so that the user gets a clear error rather than a stream . * of "command cate is unrecognized" messages. . *---------------------------------------------------------------- . . if c(stata_version) < 19 { . di as error "" . di as error "============================================================ > " . di as error " ERROR: this script requires Stata 19 or later." . di as error " Detected Stata version: " c(stata_version) . di as error " The -cate- command was introduced in Stata 19." . di as error "============================================================ > " . log close . exit 198 . } . . di as text "Stata version detected: " c(stata_version) " -- OK." Stata version detected: 19.5 -- OK. . . . *================================================================ . * Section 1: Data import, labeling, and globals . *================================================================ . * . * Import the simulated resource curse dataset. 3,000 observations . * across 300 districts in 8 fictional countries over 10 years. . * The treatment has 4 levels: 0 = no mining (~85% of obs), . * 1/2/3 = mining at low/medium/high mineral prices (~5% each). . * . * Two macro groups control the analysis: . * . * $catevars: variables that may drive heterogeneity in the . * treatment effect. These are the inputs to tau(x). . * . * $controls: additional variables for nuisance models only . * (country and year fixed effects for the panel). . *---------------------------------------------------------------- . . * Import from local CSV (update URL when uploaded to GitHub) . * GitHub: import delimited using "https://github.com/quarcs-lab/data-open/raw > /master/stata19/sim_resource_curse.csv", clear . import delimited using "sim_resource_curse.csv", clear (encoding automatically selected: ISO-8859-1) (18 vars, 3,000 obs) . . * Label all variables . label variable district_id "District ID (1-300)" . label variable country_id "Country ID (1-8)" . label variable year "Year (2003-2012)" . label variable treatment "Treatment group (0=none, 1=low, 2=med, 3=high)" . label variable mining "Mining district (binary)" . label variable price_index "Mineral price index" . label variable exec_constraints "Constraints on Executive (1-6)" . label variable quality_of_govt "Quality of Government (0.22-0.70)" . label variable gdp_pc "GDP per capita" . label variable elevation "Elevation (meters)" . label variable temperature "Mean temperature (Celsius)" . label variable ruggedness "Terrain ruggedness" . label variable distance_capital "Distance to capital (meters)" . label variable agri_suitability "Agricultural suitability (0-1)" . label variable population "Population" . label variable ethnic_frac "Ethnic fractionalization (0-1)" . label variable ntl_log "Log nighttime lights" . label variable conflict "Conflict event (binary)" . . * Create integer version of exec_constraints for group() . gen int exec_con = round(exec_constraints) . label variable exec_con "Executive Constraints (integer 1-6)" . . * Save as .dta for faster reloading . save "sim_resource_curse.dta", replace (file sim_resource_curse.dta not found) file sim_resource_curse.dta saved . . * Report dataset dimensions . describe, short Contains data from sim_resource_curse.dta Observations: 3,000 Variables: 19 6 May 2026 17:29 Sorted by: . . * Panel structure . codebook district_id, compact Variable Obs Unique Mean Min Max Label ------------------------------------------------------------------------------- district_id 3000 300 150.5 1 300 District ID (1-300) ------------------------------------------------------------------------------- . codebook country_id, compact Variable Obs Unique Mean Min Max Label ------------------------------------------------------------------------------- country_id 3000 8 4.23 1 8 Country ID (1-8) ------------------------------------------------------------------------------- . tab year, nofreq . . * Define variable lists as globals . global catevars exec_constraints quality_of_govt gdp_pc /// > elevation temperature ruggedness distance_capital /// > agri_suitability population ethnic_frac . . global controls i.country_id i.year . . . *================================================================ . * Section 2: Descriptive statistics . *================================================================ . * . * Summary statistics, treatment distribution, and outcomes by . * treatment group. The key features to note: . * - Treatment is highly imbalanced (~85% control) . * - Mining districts differ from non-mining districts . * - Institutional variables vary by country . *---------------------------------------------------------------- . . * Summary statistics for all key variables . tabstat ntl_log conflict exec_constraints quality_of_govt gdp_pc /// > elevation temperature ruggedness distance_capital /// > agri_suitability population ethnic_frac, /// > statistics(mean sd min max) columns(statistics) format(%9.3f) Variable | Mean SD Min Max -------------+---------------------------------------- ntl_log | -1.096 0.435 -2.503 0.265 conflict | 0.123 0.328 0.000 1.000 exec_const~s | 3.680 1.489 1.000 6.000 quality_of~t | 0.440 0.152 0.220 0.700 gdp_pc | 2198.000 1469.937 500.000 5000.000 elevation | 499.083 302.031 0.000 1357.232 temperature | 23.913 3.920 13.993 35.000 ruggedness | 24.423 17.803 0.000 76.953 distance_c~l | 2.68e+05 1.44e+05 10813.747 4.97e+05 agri_suita~y | 0.395 0.197 0.000 0.983 population | 82028.426 85186.961 4134.682 5.97e+05 ethnic_frac | 0.550 0.202 0.201 0.899 ------------------------------------------------------ . . * Treatment distribution . tab treatment, missing Treatment | group | (0=none, | 1=low, | 2=med, | 3=high) | Freq. Percent Cum. ------------+----------------------------------- 0 | 2,550 85.00 85.00 1 | 150 5.00 90.00 2 | 150 5.00 95.00 3 | 150 5.00 100.00 ------------+----------------------------------- Total | 3,000 100.00 . . * Mining share . count if treatment > 0 450 . local mining_n = r(N) . quietly count . display _newline "Mining share: " %5.1f 100*(`mining_n'/r(N)) "%" Mining share: 15.0% . . * Outcomes by treatment group . table treatment, statistic(mean ntl_log) statistic(mean conflict) /// > statistic(count ntl_log) nformat(%9.3f) ------------------------------------------------------------------------------------------------------------------------------ | Mean Number of nonmissing values | Log nighttime lights Conflict event (binary) Log nighttime lights -----------------------------------------------+------------------------------------------------------------------------------ Treatment group (0=none, 1=low, 2=med, 3=high) | 0 | -1.137 0.107 2550.000 1 | -1.028 0.180 150.000 2 | -0.930 0.180 150.000 3 | -0.615 0.280 150.000 Total | -1.096 0.123 3000.000 ------------------------------------------------------------------------------------------------------------------------------ . . . *================================================================ . * Section 3: Naive comparison vs ground truth . *================================================================ . * . * Raw difference-in-means for key contrasts. These are biased . * because mining districts differ systematically from non-mining . * districts. The ground-truth ATEs come from the data-generating . * process (known because the data are simulated). . *---------------------------------------------------------------- . . display as text _newline "=== Naive Difference-in-Means (biased) ===" === Naive Difference-in-Means (biased) === . display as text "Comparison" _col(20) "NTL diff" _col(35) "Ground Truth" Comparison NTL diff Ground Truth . display as text "{hline 50}" -------------------------------------------------- . . * 1-0: mining vs no mining . quietly summarize ntl_log if treatment == 1 . local m1 = r(mean) . quietly summarize ntl_log if treatment == 0 . local m0 = r(mean) . display as result "1 vs 0" _col(20) %7.4f (`m1' - `m0') _col(35) "0.25" 1 vs 0 0.1092 0.25 . . * 2-0: medium mining vs no mining . quietly summarize ntl_log if treatment == 2 . local m2 = r(mean) . display as result "2 vs 0" _col(20) %7.4f (`m2' - `m0') _col(35) "0.30" 2 vs 0 0.2077 0.30 . . * 3-0: high mining vs no mining . quietly summarize ntl_log if treatment == 3 . local m3 = r(mean) . display as result "3 vs 0" _col(20) %7.4f (`m3' - `m0') _col(35) "0.55" 3 vs 0 0.5227 0.55 . . * 2-1: medium vs low prices . display as result "2 vs 1" _col(20) %7.4f (`m2' - `m1') _col(35) "0.05" 2 vs 1 0.0985 0.05 . . * 3-1: high vs low prices . display as result "3 vs 1" _col(20) %7.4f (`m3' - `m1') _col(35) "0.30" 3 vs 1 0.4135 0.30 . . * 3-2: high vs medium prices . display as result "3 vs 2" _col(20) %7.4f (`m3' - `m2') _col(35) "0.25" 3 vs 2 0.3150 0.25 . . display as text "{hline 50}" -------------------------------------------------- . . . *================================================================ . * Section 4: ATE -- NTL 1 vs 0 (PO + AIPW) . *================================================================ . * . * Mining vs No Mining on nighttime lights. This is the most . * important contrast for Finding 1. We show both estimators: . * PO = partialing-out (partial linear model) . * AIPW = augmented inverse-probability weighting (doubly robust) . * . * Ground truth ATE: 0.25 . *---------------------------------------------------------------- . . *--- PO estimator --- . preserve . keep if treatment == 1 | treatment == 0 (300 observations deleted) . gen byte treat_1v0 = (treatment == 1) . label define treat_1v0_lbl 0 "No mining" 1 "Mining (low price)" . label values treat_1v0 treat_1v0_lbl . . display _newline as text "=== NTL: Mining vs No Mining (1-0) --- PO Estimator > ===" === NTL: Mining vs No Mining (1-0) --- PO Estimator === . display as text "N = " _N " observations" N = 2700 observations . . cate po (ntl_log $catevars) (treat_1v0), /// > controls($controls) /// > rseed(12345) xfolds(5) /// > omethod(rforest) tmethod(rforest) Cross-fit fold 1 of 5 ... Performing random forest for outcome ntl_log ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 2 of 5 ... Performing random forest for outcome ntl_log ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 3 of 5 ... Performing random forest for outcome ntl_log ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 4 of 5 ... Performing random forest for outcome ntl_log ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 5 of 5 ... Performing random forest for outcome ntl_log ... Performing random forest for treatment treat_1v0 ... Performing random forest for IATE ... Estimating AIPW scores ... Estimating ATE ... Conditional average treatment effects Number of observations = 2,700 Estimator: Partialing out Number of folds in cross-fit = 5 Outcome model: Random forest Number of outcome controls = 28 Treatment model: Random forest Number of treatment controls = 28 CATE model: Random forest Number of CATE variables = 10 ------------------------------------------------------------------------------ | Robust ntl_log | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- ATE | treat_1v0 | (Mining ..) | vs | No mining) | .1936814 .0097428 19.88 0.000 .1745858 .212777 -------------+---------------------------------------------------------------- POmean | treat_1v0 | No mining | -1.142413 .0079236 -144.18 0.000 -1.157943 -1.126883 ------------------------------------------------------------------------------ . . estimates store po_ntl_1v0 . restore . . *--- AIPW estimator --- . preserve . keep if treatment == 1 | treatment == 0 (300 observations deleted) . gen byte treat_1v0 = (treatment == 1) . label define treat_1v0_lbl 0 "No mining" 1 "Mining (low price)" . label values treat_1v0 treat_1v0_lbl . . display _newline as text "=== NTL: Mining vs No Mining (1-0) --- AIPW Estimat > or ===" === NTL: Mining vs No Mining (1-0) --- AIPW Estimator === . . cate aipw (ntl_log $catevars) (treat_1v0), /// > controls($controls) /// > rseed(12345) xfolds(5) /// > omethod(rforest) tmethod(rforest) Cross-fit fold 1 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 2 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 3 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 4 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 5 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Estimating AIPW scores ... Estimating random forest for IATE ... Estimating ATE ... Conditional average treatment effects Number of observations = 2,700 Estimator: Augmented IPW Number of folds in cross-fit = 5 Outcome model: Random forest Number of outcome controls = 28 Treatment model: Random forest Number of treatment controls = 28 CATE model: Random forest Number of CATE variables = 10 ------------------------------------------------------------------------------ | Robust ntl_log | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- ATE | treat_1v0 | (Mining ..) | vs | No mining) | .1489842 .0105686 14.10 0.000 .1282701 .1696983 -------------+---------------------------------------------------------------- POmean | treat_1v0 | No mining | -1.142416 .0079187 -144.27 0.000 -1.157936 -1.126896 ------------------------------------------------------------------------------ . . estimates store aipw_ntl_1v0 . restore . . . *================================================================ . * Section 5: ATE -- NTL 3 vs 1 (PO + AIPW) . *================================================================ . * . * High vs Low mineral prices on NTL. Key contrast for Finding 2 . * (non-linear price effects). Ground truth ATE: 0.30 . * . * Note: this uses only mining districts (~300 obs total), . * so estimates will be noisier than the 1-0 comparison. . *---------------------------------------------------------------- . . *--- PO estimator --- . preserve . keep if treatment == 3 | treatment == 1 (2,700 observations deleted) . gen byte treat_3v1 = (treatment == 3) . label define treat_3v1_lbl 0 "Low price" 1 "High price" . label values treat_3v1 treat_3v1_lbl . . display _newline as text "=== NTL: High vs Low Prices (3-1) --- PO Estimator > ===" === NTL: High vs Low Prices (3-1) --- PO Estimator === . display as text "N = " _N " observations (mining districts only)" N = 300 observations (mining districts only) . . cate po (ntl_log $catevars) (treat_3v1), /// > controls($controls) /// > rseed(12345) xfolds(5) /// > omethod(rforest) tmethod(rforest) Cross-fit fold 1 of 5 ... Performing random forest for outcome ntl_log ... Performing random forest for treatment treat_3v1 ... Cross-fit fold 2 of 5 ... Performing random forest for outcome ntl_log ... Performing random forest for treatment treat_3v1 ... Cross-fit fold 3 of 5 ... Performing random forest for outcome ntl_log ... Performing random forest for treatment treat_3v1 ... Cross-fit fold 4 of 5 ... Performing random forest for outcome ntl_log ... Performing random forest for treatment treat_3v1 ... Cross-fit fold 5 of 5 ... Performing random forest for outcome ntl_log ... Performing random forest for treatment treat_3v1 ... Performing random forest for IATE ... Estimating AIPW scores ... Estimating ATE ... Conditional average treatment effects Number of observations = 300 Estimator: Partialing out Number of folds in cross-fit = 5 Outcome model: Random forest Number of outcome controls = 28 Treatment model: Random forest Number of treatment controls = 28 CATE model: Random forest Number of CATE variables = 10 ------------------------------------------------------------------------------ | Robust ntl_log | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- ATE | treat_3v1 | (High price | vs | Low price) | .5945629 .0313138 18.99 0.000 .5331891 .6559368 -------------+---------------------------------------------------------------- POmean | treat_3v1 | Low price | -1.12839 .0280085 -40.29 0.000 -1.183285 -1.073494 ------------------------------------------------------------------------------ . . estimates store po_ntl_3v1 . restore . . *--- AIPW estimator --- . preserve . keep if treatment == 3 | treatment == 1 (2,700 observations deleted) . gen byte treat_3v1 = (treatment == 3) . label define treat_3v1_lbl 0 "Low price" 1 "High price" . label values treat_3v1 treat_3v1_lbl . . display _newline as text "=== NTL: High vs Low Prices (3-1) --- AIPW Estimato > r ===" === NTL: High vs Low Prices (3-1) --- AIPW Estimator === . . cate aipw (ntl_log $catevars) (treat_3v1), /// > controls($controls) /// > rseed(12345) xfolds(5) /// > omethod(rforest) tmethod(rforest) Cross-fit fold 1 of 5 ... Estimating random forest for outcome ntl_log if treat_3v1 = 0 ... Estimating random forest for outcome ntl_log if treat_3v1 = 1 ... Performing random forest for treatment treat_3v1 ... Cross-fit fold 2 of 5 ... Estimating random forest for outcome ntl_log if treat_3v1 = 0 ... Estimating random forest for outcome ntl_log if treat_3v1 = 1 ... Performing random forest for treatment treat_3v1 ... Cross-fit fold 3 of 5 ... Estimating random forest for outcome ntl_log if treat_3v1 = 0 ... Estimating random forest for outcome ntl_log if treat_3v1 = 1 ... Performing random forest for treatment treat_3v1 ... Cross-fit fold 4 of 5 ... Estimating random forest for outcome ntl_log if treat_3v1 = 0 ... Estimating random forest for outcome ntl_log if treat_3v1 = 1 ... Performing random forest for treatment treat_3v1 ... Cross-fit fold 5 of 5 ... Estimating random forest for outcome ntl_log if treat_3v1 = 0 ... Estimating random forest for outcome ntl_log if treat_3v1 = 1 ... Performing random forest for treatment treat_3v1 ... Estimating AIPW scores ... Estimating random forest for IATE ... Estimating ATE ... Conditional average treatment effects Number of observations = 300 Estimator: Augmented IPW Number of folds in cross-fit = 5 Outcome model: Random forest Number of outcome controls = 28 Treatment model: Random forest Number of treatment controls = 28 CATE model: Random forest Number of CATE variables = 10 ------------------------------------------------------------------------------ | Robust ntl_log | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- ATE | treat_3v1 | (High price | vs | Low price) | .4052631 .0254935 15.90 0.000 .3552968 .4552293 -------------+---------------------------------------------------------------- POmean | treat_3v1 | Low price | -1.029871 .0240718 -42.78 0.000 -1.077051 -.9826917 ------------------------------------------------------------------------------ . . estimates store aipw_ntl_3v1 . restore . . . *================================================================ . * Section 6: ATE -- NTL remaining comparisons (AIPW only) . *================================================================ . * . * Four remaining NTL contrasts using AIPW with default lasso . * methods (faster than rforest on smaller subsamples). . * . * Ground truths: 2-0 = 0.30, 3-0 = 0.55, 2-1 = 0.05, 3-2 = 0.25 . *---------------------------------------------------------------- . . *--- NTL: 2 vs 0 (medium mining vs no mining) --- . preserve . keep if treatment == 2 | treatment == 0 (300 observations deleted) . gen byte treat_2v0 = (treatment == 2) . . display _newline as text "=== NTL: Medium Mining vs No Mining (2-0) ===" === NTL: Medium Mining vs No Mining (2-0) === . display as text "N = " _N N = 2700 . . cate aipw (ntl_log $catevars) (treat_2v0), /// > controls($controls) /// > rseed(12345) xfolds(5) Cross-fit fold 1 of 5 ... Estimating lasso for outcome ntl_log if treat_2v0 = 0 ... Estimating lasso for outcome ntl_log if treat_2v0 = 1 ... Performing lasso for treatment treat_2v0 ... Cross-fit fold 2 of 5 ... Estimating lasso for outcome ntl_log if treat_2v0 = 0 ... Estimating lasso for outcome ntl_log if treat_2v0 = 1 ... Performing lasso for treatment treat_2v0 ... Cross-fit fold 3 of 5 ... Estimating lasso for outcome ntl_log if treat_2v0 = 0 ... Estimating lasso for outcome ntl_log if treat_2v0 = 1 ... Performing lasso for treatment treat_2v0 ... Cross-fit fold 4 of 5 ... Estimating lasso for outcome ntl_log if treat_2v0 = 0 ... Estimating lasso for outcome ntl_log if treat_2v0 = 1 ... Performing lasso for treatment treat_2v0 ... Cross-fit fold 5 of 5 ... Estimating lasso for outcome ntl_log if treat_2v0 = 0 ... Estimating lasso for outcome ntl_log if treat_2v0 = 1 ... Performing lasso for treatment treat_2v0 ... Estimating AIPW scores ... Estimating random forest for IATE ... Estimating ATE ... Conditional average treatment effects Number of observations = 2,700 Estimator: Augmented IPW Number of folds in cross-fit = 5 Outcome model: Linear lasso Number of outcome controls = 28 Treatment model: Logit lasso Number of treatment controls = 28 CATE model: Random forest Number of CATE variables = 10 ------------------------------------------------------------------------------ | Robust ntl_log | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- ATE | treat_2v0 | (1 vs 0) | .2891968 .0250557 11.54 0.000 .2400886 .3383049 -------------+---------------------------------------------------------------- POmean | 0.treat_2v0 | -1.14181 .0079444 -143.73 0.000 -1.15738 -1.126239 ------------------------------------------------------------------------------ . . estimates store aipw_ntl_2v0 . restore . . *--- NTL: 3 vs 0 (high mining vs no mining) --- . preserve . keep if treatment == 3 | treatment == 0 (300 observations deleted) . gen byte treat_3v0 = (treatment == 3) . . display _newline as text "=== NTL: High Mining vs No Mining (3-0) ===" === NTL: High Mining vs No Mining (3-0) === . display as text "N = " _N N = 2700 . . cate aipw (ntl_log $catevars) (treat_3v0), /// > controls($controls) /// > rseed(12345) xfolds(5) Cross-fit fold 1 of 5 ... Estimating lasso for outcome ntl_log if treat_3v0 = 0 ... Estimating lasso for outcome ntl_log if treat_3v0 = 1 ... Performing lasso for treatment treat_3v0 ... Cross-fit fold 2 of 5 ... Estimating lasso for outcome ntl_log if treat_3v0 = 0 ... Estimating lasso for outcome ntl_log if treat_3v0 = 1 ... Performing lasso for treatment treat_3v0 ... Cross-fit fold 3 of 5 ... Estimating lasso for outcome ntl_log if treat_3v0 = 0 ... Estimating lasso for outcome ntl_log if treat_3v0 = 1 ... Performing lasso for treatment treat_3v0 ... Cross-fit fold 4 of 5 ... Estimating lasso for outcome ntl_log if treat_3v0 = 0 ... Estimating lasso for outcome ntl_log if treat_3v0 = 1 ... Performing lasso for treatment treat_3v0 ... Cross-fit fold 5 of 5 ... Estimating lasso for outcome ntl_log if treat_3v0 = 0 ... Estimating lasso for outcome ntl_log if treat_3v0 = 1 ... Performing lasso for treatment treat_3v0 ... Estimating AIPW scores ... Estimating random forest for IATE ... Estimating ATE ... Conditional average treatment effects Number of observations = 2,700 Estimator: Augmented IPW Number of folds in cross-fit = 5 Outcome model: Linear lasso Number of outcome controls = 28 Treatment model: Logit lasso Number of treatment controls = 28 CATE model: Random forest Number of CATE variables = 10 ------------------------------------------------------------------------------ | Robust ntl_log | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- ATE | treat_3v0 | (1 vs 0) | .6111885 .0250606 24.39 0.000 .5620707 .6603063 -------------+---------------------------------------------------------------- POmean | 0.treat_3v0 | -1.137586 .0079516 -143.06 0.000 -1.153171 -1.122001 ------------------------------------------------------------------------------ . . estimates store aipw_ntl_3v0 . restore . . *--- NTL: 2 vs 1 (medium vs low prices) --- . preserve . keep if treatment == 2 | treatment == 1 (2,700 observations deleted) . gen byte treat_2v1 = (treatment == 2) . . display _newline as text "=== NTL: Medium vs Low Prices (2-1) ===" === NTL: Medium vs Low Prices (2-1) === . display as text "N = " _N " (within-mining comparison)" N = 300 (within-mining comparison) . . cate aipw (ntl_log $catevars) (treat_2v1), /// > controls($controls) /// > rseed(12345) xfolds(5) /// > pstolerance(1e-6) Cross-fit fold 1 of 5 ... Estimating lasso for outcome ntl_log if treat_2v1 = 0 ... Estimating lasso for outcome ntl_log if treat_2v1 = 1 ... Performing lasso for treatment treat_2v1 ... Cross-fit fold 2 of 5 ... Estimating lasso for outcome ntl_log if treat_2v1 = 0 ... Estimating lasso for outcome ntl_log if treat_2v1 = 1 ... Performing lasso for treatment treat_2v1 ... Cross-fit fold 3 of 5 ... Estimating lasso for outcome ntl_log if treat_2v1 = 0 ... Estimating lasso for outcome ntl_log if treat_2v1 = 1 ... Performing lasso for treatment treat_2v1 ... Cross-fit fold 4 of 5 ... Estimating lasso for outcome ntl_log if treat_2v1 = 0 ... Estimating lasso for outcome ntl_log if treat_2v1 = 1 ... Performing lasso for treatment treat_2v1 ... Cross-fit fold 5 of 5 ... Estimating lasso for outcome ntl_log if treat_2v1 = 0 ... Estimating lasso for outcome ntl_log if treat_2v1 = 1 ... Performing lasso for treatment treat_2v1 ... Estimating AIPW scores ... Estimating random forest for IATE ... Estimating ATE ... Conditional average treatment effects Number of observations = 300 Estimator: Augmented IPW Number of folds in cross-fit = 5 Outcome model: Linear lasso Number of outcome controls = 28 Treatment model: Logit lasso Number of treatment controls = 28 CATE model: Random forest Number of CATE variables = 10 ------------------------------------------------------------------------------ | Robust ntl_log | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- ATE | treat_2v1 | (1 vs 0) | -.0112177 .0883033 -0.13 0.899 -.1842889 .1618535 -------------+---------------------------------------------------------------- POmean | 0.treat_2v1 | -1.01681 .0310857 -32.71 0.000 -1.077737 -.9558831 ------------------------------------------------------------------------------ . . estimates store aipw_ntl_2v1 . restore . . *--- NTL: 3 vs 2 (high vs medium prices) --- . preserve . keep if treatment == 3 | treatment == 2 (2,700 observations deleted) . gen byte treat_3v2 = (treatment == 3) . . display _newline as text "=== NTL: High vs Medium Prices (3-2) ===" === NTL: High vs Medium Prices (3-2) === . display as text "N = " _N " (within-mining comparison)" N = 300 (within-mining comparison) . . cate aipw (ntl_log $catevars) (treat_3v2), /// > controls($controls) /// > rseed(12345) xfolds(5) /// > pstolerance(1e-6) Cross-fit fold 1 of 5 ... Estimating lasso for outcome ntl_log if treat_3v2 = 0 ... Estimating lasso for outcome ntl_log if treat_3v2 = 1 ... Performing lasso for treatment treat_3v2 ... Cross-fit fold 2 of 5 ... Estimating lasso for outcome ntl_log if treat_3v2 = 0 ... Estimating lasso for outcome ntl_log if treat_3v2 = 1 ... Performing lasso for treatment treat_3v2 ... treatment 0 has 1 propensity scores less than 1.00e-06 Treatment overlap assumption has been violated; use option osample() to identify the overlap violators. r(498); end of do-file r(498); ------------------------------------------------------------------------------- name: log: /Users/carlosmendez/Documents/GitHub/starter-academic-v501/content > /post/stata_cate2/analysis.log log type: text opened on: 6 May 2026, 17:38:04 . . use "sim_resource_curse.dta", clear . . global catevars exec_constraints quality_of_govt gdp_pc /// > elevation temperature ruggedness distance_capital /// > agri_suitability population ethnic_frac . global controls i.country_id i.year . . *--- NTL: 3 vs 2 (PO -- AIPW fails on overlap) --- . preserve . keep if treatment == 3 | treatment == 2 (2,700 observations deleted) . gen byte treat_3v2 = (treatment == 3) . display _newline as text "=== NTL: High vs Medium Prices (3-2) ===" === NTL: High vs Medium Prices (3-2) === . display as text "N = " _N " (within-mining comparison)" N = 300 (within-mining comparison) . cate po (ntl_log $catevars) (treat_3v2), /// > controls($controls) rseed(12345) xfolds(5) Cross-fit fold 1 of 5 ... Performing lasso for outcome ntl_log ... Performing lasso for treatment treat_3v2 ... Cross-fit fold 2 of 5 ... Performing lasso for outcome ntl_log ... Performing lasso for treatment treat_3v2 ... treatment 0 has 1 propensity scores less than 1.00e-05 Treatment overlap assumption has been violated; use option osample() to identify the overlap violators. r(498); end of do-file r(498); ------------------------------------------------------------------------------- name: log: /Users/carlosmendez/Documents/GitHub/starter-academic-v501/content > /post/stata_cate2/analysis.log log type: text opened on: 6 May 2026, 17:39:03 . . use "sim_resource_curse.dta", clear . . global catevars exec_constraints quality_of_govt gdp_pc /// > elevation temperature ruggedness distance_capital /// > agri_suitability population ethnic_frac . global controls i.country_id i.year . . *--- NTL: 3 vs 2 (PO with relaxed overlap tolerance) --- . preserve . keep if treatment == 3 | treatment == 2 (2,700 observations deleted) . gen byte treat_3v2 = (treatment == 3) . display _newline as text "=== NTL: High vs Medium Prices (3-2) ===" === NTL: High vs Medium Prices (3-2) === . display as text "N = " _N " (within-mining comparison)" N = 300 (within-mining comparison) . cate po (ntl_log $catevars) (treat_3v2), /// > controls($controls) rseed(12345) xfolds(5) /// > pstolerance(1e-8) Cross-fit fold 1 of 5 ... Performing lasso for outcome ntl_log ... Performing lasso for treatment treat_3v2 ... Cross-fit fold 2 of 5 ... Performing lasso for outcome ntl_log ... Performing lasso for treatment treat_3v2 ... Cross-fit fold 3 of 5 ... Performing lasso for outcome ntl_log ... Performing lasso for treatment treat_3v2 ... Cross-fit fold 4 of 5 ... Performing lasso for outcome ntl_log ... Performing lasso for treatment treat_3v2 ... Cross-fit fold 5 of 5 ... Performing lasso for outcome ntl_log ... Performing lasso for treatment treat_3v2 ... Performing random forest for IATE ... Estimating AIPW scores ... Estimating ATE ... Conditional average treatment effects Number of observations = 300 Estimator: Partialing out Number of folds in cross-fit = 5 Outcome model: Linear lasso Number of outcome controls = 28 Treatment model: Logit lasso Number of treatment controls = 28 CATE model: Random forest Number of CATE variables = 10 ------------------------------------------------------------------------------ | Robust ntl_log | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- ATE | treat_3v2 | (1 vs 0) | -43824.93 43752.1 -1.00 0.317 -129577.5 41927.6 -------------+---------------------------------------------------------------- POmean | 0.treat_3v2 | 43824.29 43752.1 1.00 0.317 -41928.24 129576.8 ------------------------------------------------------------------------------ . estimates store aipw_ntl_3v2 . restore . . *--- Section 7: Conflict --- . preserve . keep if treatment == 1 | treatment == 0 (300 observations deleted) . gen byte treat_1v0 = (treatment == 1) . display _newline as text "=== Conflict: Mining vs No Mining (1-0) --- PO Esti > mator ===" === Conflict: Mining vs No Mining (1-0) --- PO Estimator === . cate po (conflict $catevars) (treat_1v0), /// > controls($controls) rseed(12345) xfolds(5) /// > omethod(rforest) tmethod(rforest) Cross-fit fold 1 of 5 ... Performing random forest for outcome conflict ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 2 of 5 ... Performing random forest for outcome conflict ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 3 of 5 ... Performing random forest for outcome conflict ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 4 of 5 ... Performing random forest for outcome conflict ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 5 of 5 ... Performing random forest for outcome conflict ... Performing random forest for treatment treat_1v0 ... Performing random forest for IATE ... Estimating AIPW scores ... Estimating ATE ... Conditional average treatment effects Number of observations = 2,700 Estimator: Partialing out Number of folds in cross-fit = 5 Outcome model: Random forest Number of outcome controls = 28 Treatment model: Random forest Number of treatment controls = 28 CATE model: Random forest Number of CATE variables = 10 ------------------------------------------------------------------------------ | Robust conflict | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- ATE | treat_1v0 | (1 vs 0) | .0630853 .0130031 4.85 0.000 .0375997 .0885709 -------------+---------------------------------------------------------------- POmean | 0.treat_1v0 | .1073193 .0060727 17.67 0.000 .0954169 .1192216 ------------------------------------------------------------------------------ . estimates store po_conf_1v0 . restore . . preserve . keep if treatment == 1 | treatment == 0 (300 observations deleted) . gen byte treat_1v0 = (treatment == 1) . display _newline as text "=== Conflict: Mining vs No Mining (1-0) --- AIPW Es > timator ===" === Conflict: Mining vs No Mining (1-0) --- AIPW Estimator === . cate aipw (conflict $catevars) (treat_1v0), /// > controls($controls) rseed(12345) xfolds(5) /// > omethod(rforest) tmethod(rforest) Cross-fit fold 1 of 5 ... Estimating random forest for outcome conflict if treat_1v0 = 0 ... Estimating random forest for outcome conflict if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 2 of 5 ... Estimating random forest for outcome conflict if treat_1v0 = 0 ... Estimating random forest for outcome conflict if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 3 of 5 ... Estimating random forest for outcome conflict if treat_1v0 = 0 ... Estimating random forest for outcome conflict if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 4 of 5 ... Estimating random forest for outcome conflict if treat_1v0 = 0 ... Estimating random forest for outcome conflict if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 5 of 5 ... Estimating random forest for outcome conflict if treat_1v0 = 0 ... Estimating random forest for outcome conflict if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Estimating AIPW scores ... Estimating random forest for IATE ... Estimating ATE ... Conditional average treatment effects Number of observations = 2,700 Estimator: Augmented IPW Number of folds in cross-fit = 5 Outcome model: Random forest Number of outcome controls = 28 Treatment model: Random forest Number of treatment controls = 28 CATE model: Random forest Number of CATE variables = 10 ------------------------------------------------------------------------------ | Robust conflict | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- ATE | treat_1v0 | (1 vs 0) | .0659767 .0122036 5.41 0.000 .042058 .0898954 -------------+---------------------------------------------------------------- POmean | 0.treat_1v0 | .1075852 .0060708 17.72 0.000 .0956866 .1194838 ------------------------------------------------------------------------------ . estimates store aipw_conf_1v0 . restore . . * Conflict remaining -- use capture for overlap fallback . local comparisons "2_0 3_0 2_1 3_1 3_2" . foreach comp of local comparisons { 2. local t_hi = substr("`comp'", 1, 1) 3. local t_lo = substr("`comp'", 3, 1) 4. preserve 5. keep if treatment == `t_hi' | treatment == `t_lo' 6. gen byte treat_bin = (treatment == `t_hi') 7. display _newline as text "=== Conflict: Treatment `t_hi' vs `t_lo' === > " 8. display as text "N = " _N 9. capture noisily { 10. quietly cate aipw (conflict $catevars) (treat_bin), /// > controls($controls) rseed(12345) xfolds(5) /// > pstolerance(1e-8) 11. matrix b = e(b) 12. matrix V = e(V) 13. display as result " ATE = " %7.4f b[1,1] /// > " SE = " %7.4f sqrt(V[1,1]) 14. estimates store aipw_conf_`t_hi'v`t_lo' 15. } 16. if _rc != 0 { 17. display as text " -> Overlap violation, switching to PO estimator > " 18. capture noisily { 19. quietly cate po (conflict $catevars) (treat_bin), /// > controls($controls) rseed(12345) xfolds(5) /// > pstolerance(1e-8) 20. matrix b = e(b) 21. matrix V = e(V) 22. display as result " ATE (PO) = " %7.4f b[1,1] /// > " SE = " %7.4f sqrt(V[1,1]) 23. estimates store aipw_conf_`t_hi'v`t_lo' 24. } 25. if _rc != 0 { 26. display as text " -> Both estimators failed. Skipping." 27. } 28. } 29. restore 30. } (300 observations deleted) === Conflict: Treatment 2 vs 0 === N = 2700 ATE = 0.0728 SE = 0.0330 (300 observations deleted) === Conflict: Treatment 3 vs 0 === N = 2700 ATE = 0.1586 SE = 0.0380 (2,700 observations deleted) === Conflict: Treatment 2 vs 1 === N = 300 ATE = -0.0677 SE = 0.0497 (2,700 observations deleted) === Conflict: Treatment 3 vs 1 === N = 300 ATE = 0.1126 SE = 0.0293 (2,700 observations deleted) === Conflict: Treatment 3 vs 2 === N = 300 ATE = 3.5e+04 SE = 3.5e+04 . . *--- Section 8: ATE summary table --- . display _newline as text "{hline 70}" ---------------------------------------------------------------------- . display as text "SUMMARY: Average Treatment Effects (NTL Outcome)" SUMMARY: Average Treatment Effects (NTL Outcome) . display as text "{hline 70}" ---------------------------------------------------------------------- . display as text "Contrast" _col(15) "ATE" _col(30) "SE" _col(42) "Ground Trut > h" Contrast ATE SE Ground Truth . display as text "{hline 70}" ---------------------------------------------------------------------- . local comps "1v0 2v0 3v0 2v1 3v1 3v2" . local gts "0.25 0.30 0.55 0.05 0.30 0.25" . local i = 1 . foreach comp of local comps { 2. local gt : word `i' of `gts' 3. capture quietly estimates restore aipw_ntl_`comp' 4. if _rc == 0 { 5. matrix b = e(b) 6. matrix V = e(V) 7. local ate = b[1,1] 8. local se = sqrt(V[1,1]) 9. display as result "`comp'" _col(15) %7.4f `ate' _col(30) %7.4f `se > ' _col(42) "`gt'" 10. } 11. else { 12. display as result "`comp'" _col(15) "(not stored)" _col(42) "`gt'" 13. } 14. local ++i 15. } 1v0 (not stored) 0.25 2v0 (not stored) 0.30 3v0 (not stored) 0.55 2v1 (not stored) 0.05 3v1 (not stored) 0.30 3v2 -4.4e+04 4.4e+04 0.25 . display as text "{hline 70}" ---------------------------------------------------------------------- . . *--- Section 9: GATEs by exec_con --- . preserve . keep if treatment == 1 | treatment == 0 (300 observations deleted) . gen byte treat_1v0 = (treatment == 1) . display _newline as text "=== GATEs: NTL Mining Effect (1-0) by Executive Con > straints ===" === GATEs: NTL Mining Effect (1-0) by Executive Constraints === . cate aipw (ntl_log $catevars) (treat_1v0), /// > controls($controls) group(exec_con) /// > rseed(12345) xfolds(5) /// > omethod(rforest) tmethod(rforest) Cross-fit fold 1 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 2 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 3 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 4 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 5 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Estimating AIPW scores ... Estimating random forest for IATE ... Estimating GATE ... Conditional average treatment effects Number of observations = 2,700 Estimator: Augmented IPW Number of folds in cross-fit = 5 Outcome model: Random forest Number of outcome controls = 34 Treatment model: Random forest Number of treatment controls = 34 CATE model: Random forest Number of CATE variables = 16 ------------------------------------------------------------------------------ | Robust ntl_log | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- GATE | exec_con | 1 | .2748407 .0403765 6.81 0.000 .1957042 .3539772 2 | .3155337 .0204714 15.41 0.000 .2754106 .3556569 3 | .1674459 .020837 8.04 0.000 .1266061 .2082857 4 | .1131603 .0263687 4.29 0.000 .0614785 .164842 5 | .0998745 .0296118 3.37 0.001 .0418364 .1579127 6 | .0508165 .0220009 2.31 0.021 .0076955 .0939374 -------------+---------------------------------------------------------------- ATE | treat_1v0 | (1 vs 0) | .1517508 .0111077 13.66 0.000 .12998 .1735216 -------------+---------------------------------------------------------------- POmean | 0.treat_1v0 | -1.142328 .0079175 -144.28 0.000 -1.157846 -1.12681 ------------------------------------------------------------------------------ . categraph gateplot . graph export "stata_cate2_gate_ntl_1v0_exec.png", replace width(1200) (file stata_cate2_gate_ntl_1v0_exec.png not found) file stata_cate2_gate_ntl_1v0_exec.png written in PNG format . estat gatetest Group treatment-effects heterogeneity test H0: Group average treatment effects are homogeneous ( 1) [GATE]1bn.exec_con - [GATE]2.exec_con = 0 ( 2) [GATE]1bn.exec_con - [GATE]3.exec_con = 0 ( 3) [GATE]1bn.exec_con - [GATE]4.exec_con = 0 ( 4) [GATE]1bn.exec_con - [GATE]5.exec_con = 0 ( 5) [GATE]1bn.exec_con - [GATE]6.exec_con = 0 chi2(5) = 96.90 Prob > chi2 = 0.0000 . estimates store gate_ntl_1v0_exec . restore . . preserve . keep if treatment == 3 | treatment == 1 (2,700 observations deleted) . gen byte treat_3v1 = (treatment == 3) . display _newline as text "=== GATEs: NTL Price Effect (3-1) by Executive Cons > traints ===" === GATEs: NTL Price Effect (3-1) by Executive Constraints === . cate aipw (ntl_log $catevars) (treat_3v1), /// > controls($controls) group(exec_con) /// > rseed(12345) xfolds(5) /// > omethod(rforest) tmethod(rforest) Cross-fit fold 1 of 5 ... Estimating random forest for outcome ntl_log if treat_3v1 = 0 ... Estimating random forest for outcome ntl_log if treat_3v1 = 1 ... Performing random forest for treatment treat_3v1 ... Cross-fit fold 2 of 5 ... Estimating random forest for outcome ntl_log if treat_3v1 = 0 ... Estimating random forest for outcome ntl_log if treat_3v1 = 1 ... Performing random forest for treatment treat_3v1 ... Cross-fit fold 3 of 5 ... Estimating random forest for outcome ntl_log if treat_3v1 = 0 ... Estimating random forest for outcome ntl_log if treat_3v1 = 1 ... Performing random forest for treatment treat_3v1 ... Cross-fit fold 4 of 5 ... Estimating random forest for outcome ntl_log if treat_3v1 = 0 ... Estimating random forest for outcome ntl_log if treat_3v1 = 1 ... Performing random forest for treatment treat_3v1 ... Cross-fit fold 5 of 5 ... Estimating random forest for outcome ntl_log if treat_3v1 = 0 ... Estimating random forest for outcome ntl_log if treat_3v1 = 1 ... Performing random forest for treatment treat_3v1 ... Estimating AIPW scores ... Estimating random forest for IATE ... Estimating GATE ... Conditional average treatment effects Number of observations = 300 Estimator: Augmented IPW Number of folds in cross-fit = 5 Outcome model: Random forest Number of outcome controls = 34 Treatment model: Random forest Number of treatment controls = 34 CATE model: Random forest Number of CATE variables = 16 ------------------------------------------------------------------------------ | Robust ntl_log | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- GATE | exec_con | 1 | .3211384 .0790347 4.06 0.000 .1662332 .4760437 2 | .2868582 .0726244 3.95 0.000 .1445171 .4291993 3 | .3729897 .0413647 9.02 0.000 .2919164 .4540629 4 | .5891193 .0542141 10.87 0.000 .4828616 .6953771 5 | .4870458 .0590345 8.25 0.000 .3713403 .6027514 6 | .3400699 .0596378 5.70 0.000 .2231819 .4569579 -------------+---------------------------------------------------------------- ATE | treat_3v1 | (1 vs 0) | .4062996 .0252193 16.11 0.000 .3568707 .4557284 -------------+---------------------------------------------------------------- POmean | 0.treat_3v1 | -1.03114 .0239875 -42.99 0.000 -1.078155 -.9841252 ------------------------------------------------------------------------------ . categraph gateplot . graph export "stata_cate2_gate_ntl_3v1_exec.png", replace width(1200) (file stata_cate2_gate_ntl_3v1_exec.png not found) file stata_cate2_gate_ntl_3v1_exec.png written in PNG format . estat gatetest Group treatment-effects heterogeneity test H0: Group average treatment effects are homogeneous ( 1) [GATE]1bn.exec_con - [GATE]2.exec_con = 0 ( 2) [GATE]1bn.exec_con - [GATE]3.exec_con = 0 ( 3) [GATE]1bn.exec_con - [GATE]4.exec_con = 0 ( 4) [GATE]1bn.exec_con - [GATE]5.exec_con = 0 ( 5) [GATE]1bn.exec_con - [GATE]6.exec_con = 0 chi2(5) = 18.92 Prob > chi2 = 0.0020 . estimates store gate_ntl_3v1_exec . restore . . *--- Section 10: GATEs by quality_of_govt --- . preserve . keep if treatment == 1 | treatment == 0 (300 observations deleted) . gen byte treat_1v0 = (treatment == 1) . egen qog_cat = cut(quality_of_govt), group(4) label . display _newline as text "=== GATEs: NTL Mining Effect (1-0) by Quality of Go > vernment ===" === GATEs: NTL Mining Effect (1-0) by Quality of Government === . cate aipw (ntl_log $catevars) (treat_1v0), /// > controls($controls) group(qog_cat) /// > rseed(12345) xfolds(5) /// > omethod(rforest) tmethod(rforest) Cross-fit fold 1 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 2 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 3 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 4 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 5 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Estimating AIPW scores ... Estimating random forest for IATE ... Estimating GATE ... Conditional average treatment effects Number of observations = 2,700 Estimator: Augmented IPW Number of folds in cross-fit = 5 Outcome model: Random forest Number of outcome controls = 32 Treatment model: Random forest Number of treatment controls = 32 CATE model: Random forest Number of CATE variables = 14 ------------------------------------------------------------------------------ | Robust ntl_log | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- GATE | qog_cat | .22- | .2978846 .0215258 13.84 0.000 .2556947 .3400745 .32- | .168479 .0205681 8.19 0.000 .1281663 .2087917 .42- | .1080724 .0242792 4.45 0.000 .060486 .1556589 .58- | .0728521 .0179392 4.06 0.000 .037692 .1080123 -------------+---------------------------------------------------------------- ATE | treat_1v0 | (1 vs 0) | .1504898 .0107088 14.05 0.000 .1295009 .1714786 -------------+---------------------------------------------------------------- POmean | 0.treat_1v0 | -1.14239 .007918 -144.28 0.000 -1.157909 -1.126871 ------------------------------------------------------------------------------ . categraph gateplot . graph export "stata_cate2_gate_ntl_1v0_qog.png", replace width(1200) (file stata_cate2_gate_ntl_1v0_qog.png not found) file stata_cate2_gate_ntl_1v0_qog.png written in PNG format . estat gatetest Group treatment-effects heterogeneity test H0: Group average treatment effects are homogeneous ( 1) [GATE]0bn.qog_cat - [GATE]1.qog_cat = 0 ( 2) [GATE]0bn.qog_cat - [GATE]2.qog_cat = 0 ( 3) [GATE]0bn.qog_cat - [GATE]3.qog_cat = 0 chi2(3) = 69.19 Prob > chi2 = 0.0000 . estimates store gate_ntl_1v0_qog . restore . . preserve . keep if treatment == 3 | treatment == 1 (2,700 observations deleted) . gen byte treat_3v1 = (treatment == 3) . egen qog_cat = cut(quality_of_govt), group(4) label . display _newline as text "=== GATEs: NTL Price Effect (3-1) by Quality of Gov > ernment ===" === GATEs: NTL Price Effect (3-1) by Quality of Government === . cate aipw (ntl_log $catevars) (treat_3v1), /// > controls($controls) group(qog_cat) /// > rseed(12345) xfolds(5) /// > omethod(rforest) tmethod(rforest) Cross-fit fold 1 of 5 ... Estimating random forest for outcome ntl_log if treat_3v1 = 0 ... Estimating random forest for outcome ntl_log if treat_3v1 = 1 ... Performing random forest for treatment treat_3v1 ... Cross-fit fold 2 of 5 ... Estimating random forest for outcome ntl_log if treat_3v1 = 0 ... Estimating random forest for outcome ntl_log if treat_3v1 = 1 ... Performing random forest for treatment treat_3v1 ... Cross-fit fold 3 of 5 ... Estimating random forest for outcome ntl_log if treat_3v1 = 0 ... Estimating random forest for outcome ntl_log if treat_3v1 = 1 ... Performing random forest for treatment treat_3v1 ... Cross-fit fold 4 of 5 ... Estimating random forest for outcome ntl_log if treat_3v1 = 0 ... Estimating random forest for outcome ntl_log if treat_3v1 = 1 ... Performing random forest for treatment treat_3v1 ... Cross-fit fold 5 of 5 ... Estimating random forest for outcome ntl_log if treat_3v1 = 0 ... Estimating random forest for outcome ntl_log if treat_3v1 = 1 ... Performing random forest for treatment treat_3v1 ... Estimating AIPW scores ... Estimating random forest for IATE ... Estimating GATE ... Conditional average treatment effects Number of observations = 300 Estimator: Augmented IPW Number of folds in cross-fit = 5 Outcome model: Random forest Number of outcome controls = 32 Treatment model: Random forest Number of treatment controls = 32 CATE model: Random forest Number of CATE variables = 14 ------------------------------------------------------------------------------ | Robust ntl_log | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- GATE | qog_cat | .22- | .3224114 .0784869 4.11 0.000 .1685799 .476243 .28- | .3510705 .0408896 8.59 0.000 .2709285 .4312126 .38- | .4843956 .0567094 8.54 0.000 .3732473 .5955438 .48- | .4447202 .039344 11.30 0.000 .3676074 .5218329 -------------+---------------------------------------------------------------- ATE | treat_3v1 | (1 vs 0) | .4057735 .0253689 15.99 0.000 .3560514 .4554957 -------------+---------------------------------------------------------------- POmean | 0.treat_3v1 | -1.030637 .0240117 -42.92 0.000 -1.0777 -.9835753 ------------------------------------------------------------------------------ . categraph gateplot . graph export "stata_cate2_gate_ntl_3v1_qog.png", replace width(1200) (file stata_cate2_gate_ntl_3v1_qog.png not found) file stata_cate2_gate_ntl_3v1_qog.png written in PNG format . estat gatetest Group treatment-effects heterogeneity test H0: Group average treatment effects are homogeneous ( 1) [GATE]0bn.qog_cat - [GATE]1.qog_cat = 0 ( 2) [GATE]0bn.qog_cat - [GATE]2.qog_cat = 0 ( 3) [GATE]0bn.qog_cat - [GATE]3.qog_cat = 0 chi2(3) = 5.81 Prob > chi2 = 0.1211 . estimates store gate_ntl_3v1_qog . restore . . *--- Section 11: Conflict GATEs --- . preserve . keep if treatment == 1 | treatment == 0 (300 observations deleted) . gen byte treat_1v0 = (treatment == 1) . display _newline as text "=== GATEs: Conflict Mining Effect (1-0) by Executiv > e Constraints ===" === GATEs: Conflict Mining Effect (1-0) by Executive Constraints === . cate aipw (conflict $catevars) (treat_1v0), /// > controls($controls) group(exec_con) /// > rseed(12345) xfolds(5) /// > omethod(rforest) tmethod(rforest) Cross-fit fold 1 of 5 ... Estimating random forest for outcome conflict if treat_1v0 = 0 ... Estimating random forest for outcome conflict if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 2 of 5 ... Estimating random forest for outcome conflict if treat_1v0 = 0 ... Estimating random forest for outcome conflict if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 3 of 5 ... Estimating random forest for outcome conflict if treat_1v0 = 0 ... Estimating random forest for outcome conflict if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 4 of 5 ... Estimating random forest for outcome conflict if treat_1v0 = 0 ... Estimating random forest for outcome conflict if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 5 of 5 ... Estimating random forest for outcome conflict if treat_1v0 = 0 ... Estimating random forest for outcome conflict if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Estimating AIPW scores ... Estimating random forest for IATE ... Estimating GATE ... Conditional average treatment effects Number of observations = 2,700 Estimator: Augmented IPW Number of folds in cross-fit = 5 Outcome model: Random forest Number of outcome controls = 34 Treatment model: Random forest Number of treatment controls = 34 CATE model: Random forest Number of CATE variables = 16 ------------------------------------------------------------------------------ | Robust conflict | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- GATE | exec_con | 1 | .0924768 .0440987 2.10 0.036 .0060449 .1789087 2 | .032707 .0348295 0.94 0.348 -.0355576 .1009715 3 | .0600398 .0292415 2.05 0.040 .0027275 .1173522 4 | .0486042 .0273151 1.78 0.075 -.0049326 .1021409 5 | .0643314 .0205048 3.14 0.002 .0241427 .1045202 6 | .1057752 .021425 4.94 0.000 .0637831 .1477674 -------------+---------------------------------------------------------------- ATE | treat_1v0 | (1 vs 0) | .0648653 .0122278 5.30 0.000 .0408994 .0888313 -------------+---------------------------------------------------------------- POmean | 0.treat_1v0 | .1076284 .0060716 17.73 0.000 .0957283 .1195286 ------------------------------------------------------------------------------ . categraph gateplot . graph export "stata_cate2_gate_conf_1v0_exec.png", replace width(1200) (file stata_cate2_gate_conf_1v0_exec.png not found) file stata_cate2_gate_conf_1v0_exec.png written in PNG format . estat gatetest Group treatment-effects heterogeneity test H0: Group average treatment effects are homogeneous ( 1) [GATE]1bn.exec_con - [GATE]2.exec_con = 0 ( 2) [GATE]1bn.exec_con - [GATE]3.exec_con = 0 ( 3) [GATE]1bn.exec_con - [GATE]4.exec_con = 0 ( 4) [GATE]1bn.exec_con - [GATE]5.exec_con = 0 ( 5) [GATE]1bn.exec_con - [GATE]6.exec_con = 0 chi2(5) = 5.00 Prob > chi2 = 0.4162 . estimates store gate_conf_1v0_exec . restore . . *--- Section 12: IATE diagnostics --- . preserve . keep if treatment == 1 | treatment == 0 (300 observations deleted) . gen byte treat_1v0 = (treatment == 1) . display _newline as text "=== Stata 19 Bonus Diagnostics: NTL Mining Effect ( > 1-0) ===" === Stata 19 Bonus Diagnostics: NTL Mining Effect (1-0) === . cate aipw (ntl_log exec_constraints quality_of_govt gdp_pc /// > elevation temperature ruggedness distance_capital /// > agri_suitability population ethnic_frac /// > i.exec_con) (treat_1v0), /// > controls($controls) rseed(12345) xfolds(5) /// > omethod(rforest) tmethod(rforest) Cross-fit fold 1 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 2 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 3 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 4 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Cross-fit fold 5 of 5 ... Estimating random forest for outcome ntl_log if treat_1v0 = 0 ... Estimating random forest for outcome ntl_log if treat_1v0 = 1 ... Performing random forest for treatment treat_1v0 ... Estimating AIPW scores ... Estimating random forest for IATE ... Estimating ATE ... Conditional average treatment effects Number of observations = 2,700 Estimator: Augmented IPW Number of folds in cross-fit = 5 Outcome model: Random forest Number of outcome controls = 34 Treatment model: Random forest Number of treatment controls = 34 CATE model: Random forest Number of CATE variables = 16 ------------------------------------------------------------------------------ | Robust ntl_log | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- ATE | treat_1v0 | (1 vs 0) | .1517508 .0111077 13.66 0.000 .12998 .1735216 -------------+---------------------------------------------------------------- POmean | 0.treat_1v0 | -1.142328 .0079175 -144.28 0.000 -1.157846 -1.12681 ------------------------------------------------------------------------------ . . categraph histogram (bin=34, start=-.59263562, width=.07842205) . graph export "stata_cate2_iate_histogram.png", replace width(1200) (file stata_cate2_iate_histogram.png not found) file stata_cate2_iate_histogram.png written in PNG format . estat heterogeneity Treatment-effects heterogeneity test H0: Treatment effects are homogeneous chi2(1) = 53.05 Prob > chi2 = 0.0000 . . *--- Section 13: reestimate --- . cate, reestimate group(exec_con) Estimating GATE ... Conditional average treatment effects Number of observations = 2,700 Estimator: Augmented IPW Number of folds in cross-fit = 5 Outcome model: Random forest Number of outcome controls = 34 Treatment model: Random forest Number of treatment controls = 34 CATE model: Random forest Number of CATE variables = 16 ------------------------------------------------------------------------------ | Robust ntl_log | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- GATE | exec_con | 1 | .2748407 .0403765 6.81 0.000 .1957042 .3539772 2 | .3155337 .0204714 15.41 0.000 .2754106 .3556569 3 | .1674459 .020837 8.04 0.000 .1266061 .2082857 4 | .1131603 .0263687 4.29 0.000 .0614785 .164842 5 | .0998745 .0296118 3.37 0.001 .0418364 .1579127 6 | .0508165 .0220009 2.31 0.021 .0076955 .0939374 -------------+---------------------------------------------------------------- ATE | treat_1v0 | (1 vs 0) | .1517508 .0111077 13.66 0.000 .12998 .1735216 -------------+---------------------------------------------------------------- POmean | 0.treat_1v0 | -1.142328 .0079175 -144.28 0.000 -1.157846 -1.12681 ------------------------------------------------------------------------------ . estat gatetest Group treatment-effects heterogeneity test H0: Group average treatment effects are homogeneous ( 1) [GATE]1bn.exec_con - [GATE]2.exec_con = 0 ( 2) [GATE]1bn.exec_con - [GATE]3.exec_con = 0 ( 3) [GATE]1bn.exec_con - [GATE]4.exec_con = 0 ( 4) [GATE]1bn.exec_con - [GATE]5.exec_con = 0 ( 5) [GATE]1bn.exec_con - [GATE]6.exec_con = 0 chi2(5) = 96.90 Prob > chi2 = 0.0000 . . *--- Section 14: subpopulation ATEs --- . estat ate if exec_constraints >= 4 Estimating ATE ... Treatment-effects estimation Number of obs = 1,526 ------------------------------------------------------------------------------ | Robust | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- ATE | treat_1v0 | (1 vs 0) | .0922992 .0156897 5.88 0.000 .0615479 .1230506 -------------+---------------------------------------------------------------- POmean | 0.treat_1v0 | -.9377741 .0088158 -106.37 0.000 -.9550527 -.9204955 ------------------------------------------------------------------------------ . estat ate if exec_constraints <= 2 Estimating ATE ... Treatment-effects estimation Number of obs = 558 ------------------------------------------------------------------------------ | Robust | Coefficient std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- ATE | treat_1v0 | (1 vs 0) | .2970104 .0215156 13.80 0.000 .2548407 .3391801 -------------+---------------------------------------------------------------- POmean | 0.treat_1v0 | -1.529463 .0134543 -113.68 0.000 -1.555833 -1.503093 ------------------------------------------------------------------------------ . . *--- Section 15: linear projection --- . estat projection exec_constraints quality_of_govt gdp_pc elevation temperatur > e Treatment-effects linear projection Number of obs = 2,700 F(5, 2694) = 13.95 Prob > F = 0.0000 R-squared = 0.0235 Adj R-squared = 0.0217 Root MSE = 0.5710 ------------------------------------------------------------------------------ | Robust | Coefficient std. err. t P>|t| [95% conf. interval] -------------+---------------------------------------------------------------- exec_const~s | -.026133 .0470456 -0.56 0.579 -.1183822 .0661162 quality_of~t | -.862502 .6669174 -1.29 0.196 -2.170224 .4452196 gdp_pc | .000067 .0000388 1.73 0.084 -9.06e-06 .000143 elevation | -.0001258 .0000341 -3.69 0.000 -.0001928 -.0000589 temperature | .0045969 .0022266 2.06 0.039 .0002309 .0089629 _cons | .4351266 .1043275 4.17 0.000 .2305565 .6396967 ------------------------------------------------------------------------------ . . *--- Section 16: IATE function plots --- . categraph iateplot exec_constraints Note: IATE estimated at fixed values of covariates other than exec_constraints. -------------------------------------------------- Variable | Statistic Value Type ------------+------------------------------------- agri_sui~y | mean .3898838 continuous distance~l | mean 267647.1 continuous elevation | mean 499.6838 continuous ethnic_f~c | mean .5505581 continuous exec_con | base 1 factor gdp_pc | mean 2243.333 continuous population | mean 82696.72 continuous quality_~t | mean .444837 continuous ruggedness | mean 23.96731 continuous temperat~e | mean 23.97849 continuous -------------------------------------------------- . graph export "stata_cate2_iateplot_exec.png", replace width(1200) (file stata_cate2_iateplot_exec.png not found) file stata_cate2_iateplot_exec.png written in PNG format . categraph iateplot quality_of_govt Note: IATE estimated at fixed values of covariates other than quality_of_govt. -------------------------------------------------- Variable | Statistic Value Type ------------+------------------------------------- agri_sui~y | mean .3898838 continuous distance~l | mean 267647.1 continuous elevation | mean 499.6838 continuous ethnic_f~c | mean .5505581 continuous exec_con | base 1 factor exec_con~s | mean 3.724074 continuous gdp_pc | mean 2243.333 continuous population | mean 82696.72 continuous ruggedness | mean 23.96731 continuous temperat~e | mean 23.97849 continuous -------------------------------------------------- . graph export "stata_cate2_iateplot_qog.png", replace width(1200) (file stata_cate2_iateplot_qog.png not found) file stata_cate2_iateplot_qog.png written in PNG format . . restore . . *--- Section 17: session info --- . di _newline(2) . di "============================================================" ============================================================ . di " CATE analysis complete." CATE analysis complete. . di "============================================================" ============================================================ . about StataNow/SE 19.5 for Unix (Intel 64-bit) Revision 18 Feb 2026 Copyright 1985-2025 StataCorp LLC Total usable memory: 128.00 GB Stata license: Single-user , expiring 19 Oct 2026 Serial number: 401909342834 Licensed to: Carlos Mendez Nagoya University . display _newline "Working directory: `c(pwd)'" Working directory: /Users/carlosmendez/Documents/GitHub/starter-academic-v501/c > ontent/post/stata_cate2 . display "Date: `c(current_date)'" Date: 6 May 2026 . display "Stata version: `c(stata_version)'" Stata version: 19.5 . . log close name: log: /Users/carlosmendez/Documents/GitHub/starter-academic-v501/content > /post/stata_cate2/analysis.log log type: text closed on: 6 May 2026, 17:59:54 -------------------------------------------------------------------------------