Modeling Investment Trends: A Logarithmic-Modified Markov Chain Approach

Authors

Imoh Udo Moffat^1, , James Augustine Ukpabio¹, Emmanuel Alphonsus Akpan^{2, 3, *,}

¹Department of Statistics, University of Uyo, Uyo, Nigeria

²Department of Mathematical Science, Abubakar Tafawa Balewa University, Bauchi, Nigeria

³Federal School of Medical Laboratory Technology (Science), Jos, Nigeria

^*Corresponding author. Email: eubong44@gmail.com

DOI

10.2991/jsta.d.201006.001How to use a DOI?

Keywords

Convergence; Heteroscdasticity; Logarithmic transformation; Markov chain; Stochastic process; Transition matrix

Abstract

The study aimed at stabilizing the changing variance using the logarithmic transformation to achieve a significant proportion of stability and a faster rate of convergence of the steady state transition probability in Markov chains. The traditional Markov chain and logarithmic-modified Markov chain were considered. On exploring the yearly data on the stock prices from 2015 to 2018 as obtained from the Nigerian Stock Exchange, it was found that the steady state of logarithmic-modified Markov chain converged faster than the tradition Markov chain with efficiency in tracking the correct cycles where the stock movements are trending irrespective of which cycle it starts at time zero with differences in probability values by 1.1%, 0.7%, −0.41% and −1.37% for accumulation, markup, distribution and mark-down cycles, respectively. Thus, it could be deduced that the logarithmic modification enhances the ability of the Markov chain to tract the variation of the steady state probabilities faster than the traditional counterpart.

Copyright

© 2020 The Authors. Published by Atlantis Press B.V.

Open Access

This is an open access article distributed under the CC BY-NC 4.0 license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)
View full text (HTML)

TY  - JOUR
AU  - Imoh Udo Moffat
AU  - James Augustine Ukpabio
AU  - Emmanuel Alphonsus Akpan
PY  - 2020
DA  - 2020/10/16
TI  - Modeling Investment Trends: A Logarithmic-Modified Markov Chain Approach
JO  - Journal of Statistical Theory and Applications
SP  - 439
EP  - 445
VL  - 19
IS  - 3
SN  - 2214-1766
UR  - https://doi.org/10.2991/jsta.d.201006.001
DO  - 10.2991/jsta.d.201006.001
ID  - Moffat2020
ER  -