Modeling the Effect of Population Size on Banking Transaction Channels in Nigeria: Grey Box vs Support Vector Regression

Authors

Desmond BartholomewFollow
Ngozi P. Olewuezi, Department of Statistics, School of Physical Sciences, Federal University of Technology Owerri, Imo State, Nigeria.
Chrysogonus C. Nwaigwe, Department of Statistics, School of Physical Sciences, Federal University of Technology Owerri, Imo State, Nigeria.
Felix C. Akanno, Department of Statistics, School of Physical Sciences, Federal University of Technology Owerri, Imo State, Nigeria.

Keywords

Grey-box, banking channels, Nigeria population, support vector regression, POS, USSD

Abstract

This study investigates the effect of Nigeria’s population on four selected banking transaction channels. The Nigerian projected population (2022-2027) was used as an input variable for forecasting future volumes of transactions for each channel. The results show that the Support Vector Regression (SVR) model best fits the ATM, Online, and USSD channels of transaction while the Grey-box was better for POS. The forecast results show that ATM, online, and USSD channels had their highest volume of transactions in 2023, while for POS, the highest volume was recorded in 2027. Further results indicate that online and POS transactions would dominate payment system landscape in the future. To better serve customers in the future, Nigerian banks should expand their capacity for online and POS transactions.

Publication Title

CBN Journal of Applied Statistics

Issue

1

Volume

15

First Page

1

Last Page

31

Recommended Citation

Bartholomew, Desmond; Olewuezi, Ngozi P.; Nwaigwe, Chrysogonus C.; and Akanno, Felix C. (2024) "Modeling the Effect of Population Size on Banking Transaction Channels in Nigeria: Grey Box vs Support Vector Regression," CBN Journal of Applied Statistics (JAS): Vol. 15: No. 1, Article 1.
Available at: https://dc.cbn.gov.ng/jas/vol15/iss1/1