Data for: "Portfolio Management under Multiple Regimes: Strategies that Outperform the Market" published by RAC-Revista de Administração Contemporânea

Published: 11-02-2020| Version 1 | DOI: 10.17632/jzypktvvs8.1
Contributors:
Marcelo Lewin,
Carlos Heitor Campani

Description

The files refer to the paper "Portfolio Management under Multiple Regimes: Strategies that Outperform the Market", published by RAC-Revista de Administração Contemporânea. The time series were extracted from Economatica(c) database at COPPEAD Graduate School of Business - Federal University of Rio de Janeiro in 2018. The programs were created to solve a dynamic asset allocation problem for a single and multiple regime economy. The researcher can reproduce our results with these files. The results were promissing and our objective with this research is to open field for a broader application of regime swithing models in asset allocation worldwide.

Download All

Steps to reproduce

The data file is in Excel mode. The sheet "Parameters_Start" is where the collected time series of asset prices is converted into asset classes risk premia. The sheet "Reported_Table" has the initial parameters for the optimization procedure, and "Parameters" is the sheet that MatLab writes the results: (i) parameters estimation, and (ii) probabilities. The programs were created for MatLab 2018a. Program #1 estimates the parameters for the single regime model. Program #2 estimates the parameters for the multiple regime model and writes them in Data.xlsx. Program #3 creates the parameters file, while program #4 creates the regime probabilities and also writes them in Excel. Lastly, program #5 computes the weights to solve the asset allocation problem.