Heterogeneity in the effect of federal spending on local crime: Evidence from causal forests

Heterogeneity in the Effects of Group Lending in Microfinance-The Case of Mongolia

“…Athey and Imbens (2019) applied a tree-based method to causal inference, and Wager and Athey (2018) developed the method as Causal Forest and found that the CATE predicted by the method have congruence and asymptotic normality under certain conditions. They also predict CATE in areas such as poverty and labor studies, taxation, and energy conservation, and examine the characteristics of their distributions and what attributes explain their heterogeneity (Meller 2020, Carter, Tjernstrom and Toledo 2019, De Neve et al 2019, Farbmacher, Kogel and Spindler 2019, Hoffman and Mast 2019, O'Neil and Weeks 2018. The usefulness of implementing targets for intervention through CATE has also been suggested.…”

“…De Neve et al (2019) examined the heterogeneity of the effects of nudges on tax payment rates and found that simplifying the content of documents related to tax payment procedures alone had little effect on the elderly, that deterrent messages specifying penalties for delays in paying taxes and tax increases were more effective for the young, and that households with children For households with children, simplification is more effective. Farbmacher, Kogel and Spindler (2019) examined poverty status and cognitive ability, and Hoffman and Mast (2019) examined the effect of government spending on the incidence of crime and estimate heterogeneity of the effect by region. They show that crime rates are suppressed in areas with low income and low employment growth; O'Neil and Weeks (2018) examined the effect of hourly electricity rates on electricity saving behavior and find that the intervention effect is larger for younger and more educated households with larger electricity use.…”

Causal effect of video game play on mental well-being: a quasi-experimental study among Japanese population

“…We employ a machine learning algorithm called causal forest developed by Wager and Athey (2018) [30], predicting treatment effects on each individual based on their characteristics. This method, used in recent studies (i.e., [31,32]), allows flexible, high-dimensional combinations of covariates to identify the video game effect on each individual. For this prediction, we focus on estimating conditional ATEs (E(Y 1 − Y 0 |X = x)) capturing differences in lottery winners and non-winners.…”

Differentiating Between Outlaw Motorcycle Gangs (OMCGs): Estimating the Effect of Membership of the Most Crime-Prone OMCGs on Crime Using Matching Weights

“…One-on-one (caliper) matching on the propensity score is arguably the most popular criminological application of the propensity score (Banks & Gottfredson, 2003;Bingenheimer et al, 2005;Brame et al, 2004;King et al, 2007;Leeb et al, 2007;Mocan & Tekin, 2006;Sweeten & Apel, 2007), although some studies have applied inverse probability weighting in which individuals are weighted by the inverse of their propensity score (Hoffman & Mast, 2019;Mowen & Visher, 2015;Sampson et al, 2006). Here we opt for using matching weights, a method particularly suited when the distribution of the propensity score is skewed.…”