scholarly journals Robust Inference of Conditional Average Treatment Effects Using Dimension Reduction

2023 ◽  
Ming-Yueh Huang ◽  
Shu Yang
2020 ◽  
Vol 29 (05) ◽  
pp. 2050005
Lev V. Utkin ◽  
Mikhail V. Kots ◽  
Viacheslav S. Chukanov ◽  
Andrei V. Konstantinov ◽  
Anna A. Meldo

A new meta-algorithm for estimating the conditional average treatment effects is pro-posed in the paper. The basic idea behind the algorithm is to consider a new dataset consisting of feature vectors produced by means of concatenation of examples from control and treatment groups, which are close to each other. Outcomes of new data are defined as the difference between outcomes of the corresponding examples comprising new feature vectors. The second idea is based on the assumption that the number of controls is rather large and the control outcome function is precisely determined. This assumption allows us to augment treatments by generating feature vectors which are closed to available treatments. The outcome regression function constructed on the augmented set of concatenated feature vectors can be viewed as an estimator of the conditional average treatment effects. A simple modification of the Co-learner based on the random subspace method or the feature bagging is also proposed. Various numerical simulation experiments illustrate the proposed algorithm and show its outperformance in comparison with the well-known T-learner and X-learner for several types of the control and treatment outcome functions.

2015 ◽  
Vol 33 (4) ◽  
pp. 485-505 ◽  
Jason Abrevaya ◽  
Yu-Chin Hsu ◽  
Robert P. Lieli

2021 ◽  
Alexander Olof Savi ◽  
Chris van Klaveren ◽  
Ilja Cornelisz

Effort is key in learning, evidenced by its omnipresence in both empirical findings and educational theories. At the same time, students are consistently found to avoid effort. In this study, we investigate whether limiting effort avoidance improves learning outcomes, and explore for whom this would be the case. In a large-scale computer adaptive practice system for primary education, over 150,000 participants were distributed across four conditions in which a problem-skipping option was delayed for 0, 3, 6, or 9 seconds. The results show that after a 14 week period, no average treatment effects in learning outcomes can be found between conditions. A substantive typology of students, based on the expected target mechanisms of the intervention, neither shows consistent conditional average treatment effects. Nevertheless, the substantive typology is shown to be meaningful, as the different types—toilers, skippers, and rushers—differ with respect to their learning outcomes. We argue that although the scale of the experiment suggests a precise null finding, the cumulative nature of the effect of problem skipping cautions against generalizing this finding to sustained intervention.

Sign in / Sign up

Export Citation Format

Share Document