Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/258099 
Year of Publication: 
2021
Citation: 
[Journal:] Risks [ISSN:] 2227-9091 [Volume:] 9 [Issue:] 1 [Article No.:] 9 [Publisher:] MDPI [Place:] Basel [Year:] 2021 [Pages:] 1-18
Publisher: 
MDPI, Basel
Abstract: 
Hidden Markov model (HMM) is a powerful machine-learning method for data regime detection, especially time series data. In this paper, we establish a multi-step procedure for using HMM to select stocks from the global stock market. First, the five important factors of a stock are identified and scored based on its historical performances. Second, HMM is used to predict the regimes of six global economic indicators and find the time periods in the past during which these indicators have a combination of regimes that is similar to those predicted. Then, we analyze the five stock factors of the All country world index (ACWI) in the identified time periods to assign a weighted score for each stock factor and to calculate the composite score of the five factors. Finally, we make a monthly selection of 10% of the global stocks that have the highest composite scores. This strategy is shown to outperform those relying on either ACWI, any single stock factor, or the simple average of the five stock factors.
Subjects: 
economics
economics indicators
global stocks
hidden Markov model
machine learning
regimes
stock ranking
stocks&#x2019
factors
trading
Persistent Identifier of the first edition: 
Creative Commons License: 
cc-by Logo
Document Type: 
Article
Appears in Collections:

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.