TVP-VAR frequency connectedness analysis on CPI-based monthly real return volatility of financial investment instruments

Abstract

Purpose: This study explores volatility transmission among the real returns of financial investment instruments, using the Diebold-Yilmaz approach and data from the Turkish Statistical Institute. The dataset includes monthly real return rates of instruments like Gross Interest Rate (GIR) for deposits, ingot Gold (GOLD), Istanbul Stock Exchange 100 Index (BIST-100), United States Dollar (USD), Euro (EUR), and Government Domestic Debt Instruments (GDDI) from January 2005 to April 2023. Methodology: Real return rates were adjusted using the Consumer Price Index (CPI). Absolute values of real returns served as volatility proxies. To evaluate volatility spillover among these instruments, the Time-Varying Parameter Vector Autoregressive (TVP-VAR) frequency connectedness approach was utilized. Results: The average of the Total Connectedness Index (TCI) suggests 40.37% of error variance in investment instruments is due to network connectedness, with short-term and long-term values at 33.95% and 6.41%, respectively. Dynamic TCI values spiked during events like the 2008 crisis, 2018 and 2021 exchange rate shocks, and COVID-19. USD and EUR consistently caused net volatility spillovers, GOLD in the longrun, GDDI in the short run and aggregate. GIR was most impacted by network shocks. The study also examined the Net Pairwise Connectedness Index (NPCI) to identify dominant instruments in the network. Conclusion: The findings show the interdependencies and significant roles of particular investment instruments in the transmission of volatility, offering insights for portfolio diversification and risk management.