Research projects

Project CEMAPRE internal

TitleConditional Expected time - Estimation and Inference
ParticipantsJoão Nicolau (Principal Investigator)
SummaryNicolau (2017) proposes a simple but effective non-parametric method to estimate the expected time
(ET) to cross a threshold using only two assumptions: Markovian property and stationarity. Zurkis et
al. (2021) extend the methodology to accommodate covariates. Specifically, the proposed methodology
allows us to estimate how covariates influence the time that a process takes to cross a fixed
threshold, and thus, it provides relevant insights on the potential causal relationships between
economic variables. In this project we aim to further explore the methodology of Zurkis et al.
(2021) by analyzing topics such as goodness of fit, endogeneity, robust estimation, forecasting,
structure breaks, among other topics.
* Nicolau, J. (2017). A simple nonparametric method to estimate the expected time to cross a
threshold. Statistics & Probability Letters, 123, 146-152.
* Zsurkis, G., Nicolau, J., & Rodrigues, P. M. (2021). The expected time to cross a threshold and
its determinants: a simple and flexible framework. Journal of Economic Dynamics and Control, 122,