Modelling selected stock prices at the Nairobi securities exchange using Markov chains

Dennis Cheruiyot Kiplangat *

Department of Statistics and Actuarial Science, Dedan Kimathi University of Technology, Kenya.
 
Research Article
International Journal of Science and Research Archive, 2024, 13(02), 1528–1542.
Article DOI: 10.30574/ijsra.2024.13.2.2294
Publication history: 
Received on 16 October 2024; revised on 25 November 2024; accepted on 27 November 2024
 
Abstract: 
This paper modelled four stock prices traded at the Nairobi stock exchange using Markov Chains. Markov chain is a stochastic process that has a Markovian property. The study focused primarily on the application of Markov Chain in analysing Nairobi Securities Market daily returns to describe the distribution of the market returns, cross-checking if a current market return depends on its preceding market returns and using the analysis to predict the future returns. Discrete Markov Chains were fitted to determine the probability of change of states in the returns of stock prices and to determine the extent to which the Markov model can be used to forecast returns. The idea of using Markov chains to predict the behaviour of market prices is widely used since prospective stock investors are concerned with stock price movements which might lead to an optimum investment strategy. The Markovian test carried out in the study on the daily returns showed that the returns had a Markov property. The model predicted that the share prices would either decrease to a range of minimum and lower average, decrease to a range of lower average and zero, remain unchanged, increase to a range of zero and upper average and increase to a range of upper average. The steady-state was found in the 21st, 23rd, 12th and 21st trading days for Centum, Equity Bank, East African Breweries Limited (EABL) and Kenya Airways respectively.
 
Keywords: 
Hidden Markov Model (HMM); Initial Price Offer (IPO); Markov Chain Model (MCM); Nairobi Securities Exchange (NSE); Transition Probability Matrix (TPM).
 
Full text article in PDF: 
 
Copyright information: 

Copyright © 2024 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0