Selective linear segmentation for detecting relevant parameter changes
Preprint
Abstract
Change-point processes are one flexible approach to model long time series. We propose a method to uncover which model parameter truly vary when a change-point is detected. Given a set of breakpoints, we use a penalized likelihood approach to select the best set of parameters that changes over time and we prove that the penalty function leads to a consistent selection of the true model. Estimation is carried out via the deterministic annealing expectation-maximization algorithm. Our method accounts for model selection uncertainty and associates a probability to all the possible time-varying parameter specifications. Monte Carlo simulations highlight that the method works well for many time series models including heteroskedastic processes. For a sample of 14 Hedge funds (HF) strategies, using an asset based style pricing model, we shed light on the promising ability of our method to detect the time-varying dynamics of risk exposures as well as to forecast HF returns.Keywords
All Related Versions
- Version 1, 2024-02-08, ArXiv
- Published version: Journal of Financial Econometrics, 20 (4), 762.
This publication has 63 references indexed in Scilit:
- Sparse change‐point VAR modelsJournal of Applied Econometrics, 2021
- Should I stay or should I go? A latent threshold approach to large‐scale mixture innovation modelsJournal of Applied Econometrics, 2019
- Real Estate Risk and Hedge Fund ReturnsThe Journal of Real Estate Finance and Economics, 2015
- Ancillarity-sufficiency interweaving strategy (ASIS) for boosting MCMC estimation of stochastic volatility modelsComputational Statistics & Data Analysis, 2014
- SEQUENTIAL MONTE CARLO SAMPLING FOR DSGE MODELSJournal of Applied Econometrics, 2014
- The Contribution of Structural Break Models to Forecasting Macroeconomic SeriesJournal of Applied Econometrics, 2014
- A new structural break model, with an application to Canadian inflation forecastingInternational Journal of Forecasting, 2014
- Exact posterior distributions and model selection criteria for multiple change-point detection problemsStatistics and Computing, 2011
- Simulation smoothing for state–space models: A computational efficiency analysisComputational Statistics & Data Analysis, 2011
- A Markov Chain Monte Carlo version of the genetic algorithm Differential Evolution: easy Bayesian computing for real parameter spacesStatistics and Computing, 2006