Bayesian Dynamic Conditional Correlation Model for Kogi State Financial Time Series Correlation Analysis

Abstract

A Bayesian Dynamic Conditional Correlation (DCC) model is developed to estimate time-varying correlations among financial assets in Kogi State, Nigeria, integrating univariate GARCH(1,1) models with Bayesian estimation via Markov Chain Monte Carlo (MCMC) techniques to address parameter uncertainty and data scarcity. The model incorporates regionspecific covariates, such as agricultural output, inflation, and interest rates, to capture the dynamic interplay of economic factors influencing asset correlations. It is applied to weekly price data of yam, cassava, rice, and local banking stocks from 2015 to 2024, sourced from Kogi State’s local markets and financial institutions. Results reveal significant time-varying correlations, with yam–cassava prices exhibiting high correlations (0.4–0.8) and commodity–banking stock correlations peaking during economic shocks, providing valuable insights for portfolio optimization and risk management in Kogi State’s volatile financial markets.