The present analysis aims to assess whether some models developed on US stock data taken from the S&P 500 index are applicable to non-US stocks. To conduct our analysis, we took into account the stocks constituting the Nikkei 225 index and applied them to three specific models: 1. Network-based portfolio strategy using a signature-based matrix. 2. Kurtosis-Based Factor Risk Parity - allocating Portfolio Tail Risk. 3. AlphaPortfolio - direct construction with reinforcement learning. The results obtained in this analysis lead to the conclusion that overall the models are applicable to non-US stocks, even if some adjustments are required. In addition, given the complexity of the models and the different assumptions and approaches they use, it is not possible to assess which one is the model which better performs with Japanese stocks evaluated on a period of 34 years.

Creation of optimal portfolios by applying 3 cutting-edge models to non-US stocks

NASIC, ALMA
2024/2025

Abstract

The present analysis aims to assess whether some models developed on US stock data taken from the S&P 500 index are applicable to non-US stocks. To conduct our analysis, we took into account the stocks constituting the Nikkei 225 index and applied them to three specific models: 1. Network-based portfolio strategy using a signature-based matrix. 2. Kurtosis-Based Factor Risk Parity - allocating Portfolio Tail Risk. 3. AlphaPortfolio - direct construction with reinforcement learning. The results obtained in this analysis lead to the conclusion that overall the models are applicable to non-US stocks, even if some adjustments are required. In addition, given the complexity of the models and the different assumptions and approaches they use, it is not possible to assess which one is the model which better performs with Japanese stocks evaluated on a period of 34 years.
2024
File in questo prodotto:
File Dimensione Formato  
tesi_Nasic.pdf

embargo fino al 11/07/2026

Dimensione 4.72 MB
Formato Adobe PDF
4.72 MB Adobe PDF

I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.14247/25485