High frequency correlation dynamics and day-of-the-week effect: A score-driven approach in an emerging market stock exchange

Abstract

Intraday correlation dynamics poses challenges to financial econometricians, especially in recently popular high frequency domain, due to non-synchronous trading and market microstructure noise. Traditional models fail to address the issues inherent to the nature of the data, which is riddled with noisy signals and missing values. We employ a recently developed method based on Generalized Autoregressive Score (GAS) framework and State-Space Modeling to remedy these characteristics of high frequency data and estimate intraday correlations in Turkish equity market Borsa Istanbul. Our findings reveal that average intraday conditional correlation rises as trading commences and lingers around certain altitude for some time, with the eigenvalues associated with market factor becoming progressively more dominant. An upward trend closes out the trading day on Mondays, which we attribute to the US market opening, whereas the rest of the week does not show a generalizable closing-time effect. Assessment of the findings across different market conditions and days of the week reveals elevated correlation levels in volatile markets as well as a distinguishable path for the beginning of the week. Beyond the scholarly contribution, the methodology can be used as a nowcasting tool and the findings are of interest to various parties like high-frequency traders, risk and portfolio managers and regulatory agencies in formulating their high frequency trading practices, hedging, portfolio construction schemes and margin requirements, respectively.