Mixed hitting-time (MHT) models are mixture duration models that specify durations as the first time a latent stochastic process crosses a heterogeneous threshold. They are of substantial interest because they can be applied to the analysis of optimal stopping decisions by heterogeneous agents. Examples include models of job durations, marriage durations, and the entry and exit of firms that are driven by Brownian motions and more general persistent processes.
I started this project by specifying basic single and multiple spell MHT models and discussing their economic foundations and empirical content.
With Tim Salimans and Kamlesh Kumar, I am exploring the MHT model's empirical implementation.
With Yifan Yu, I am currently exploring the specification and identification of multivariate MHT models that can be applied to the empirical analysis of optimal stopping games.