Project CEMAPRE internal
| Title | A Tail Index Estimation Approach Using a Novel Least-Squares Framework |
| Participants | João Nicolau (Principal Investigator) |
| Summary | "This project is going to focus on a novel least-squares framework for conditional tail index estimation, extending and generalizing in various directions the work of Nicolau, J., Rodrigues, P. M., & Stoykov, M. Z. (2023). Tail index estimation in the presence of covariates: Stock returns’ tail risk dynamics. Journal of Econometrics. The framework considers cross-section or time series where the response variable, governed by a Pareto distribution, is conditional on a p-dimensional vector of explanatory variables. We will establish a simple linear regression framework for the conditional tail index regression, enabling straightforward computation of parameter estimates by ordinary least squares. We intend to explore the finite and asymptotic properties of the OLS tail index estimator for Pareto and Pareto-type distributions." |