In public good games, voluntary contributions tend to start off high and decline as the game is repeated. If high contributors are matched, however, contributions tend to stay high. We propose a formalization predicting that high contributors will self-select into groups committed to charitable...

We present a simple approach to transform a deterministic numerical model, where several agents simultaneously make decisions, into a stochastic model. This approach, which builds on scenario aggregation, a numerical method developed to solve decision problems under uncertainty, is used to build...

This paper discusses the evaluation problem using observational data when the timing of treatment is an outcome of a stochastic process. We show that, without additional assumptions, it is not possible to estimate the average treatment effect and treatment on the treated. It is, however,...