Liquidity risk analysis via drawdown-based measures

Abstract

Trading volumes are key variables in determining the degree of an asset's liquidity. We examine the volume drawdown process and crash recovery measures in rolling-time windows to assess exposure to liquidity risk. The time-varying windows protect our financial indicators from the massive amount of volume transactions that characterize the opening and closing of the stock market. The empirical study is carried out for three Nasdaq-listed assets from April to September 2022. Firstly, we shape all of the volume time series using a weighted-indexed semi-Markov (WISMC) model, as well as the EGARCH and GJR models for comparisons. Next, we calculate drawdown-based risk measures on real and synthetic data, simulated from all the considered econometric models. Finally, we employ the Kullback-Leibler divergence to compare real and simulated risk indicators. Results reveal that the WISMC model reproduces all the drawdown-based risk measures better than the EGARCH and GJR models do for all the considered stocks.