Volatility modeling and prediction: the role of price impact

Abstract

In this paper, we are interested in exploring the role of price impact, derived from the order book, in modeling and predicting stock volatility. This is motivated by the market microstructure literature that examines the mechanics of price formation and its relevance to market quality. Using a comprehensive dataset of intraday bids, asks, and three levels of market depths for 148 stocks in the Shanghai Stock Exchange from 2005 to 2016, we find substantial intraday impact from incoming bid and ask limit and market orders on stock prices. More importantly, the permanent price impact at the daily level is a significant determinant of stock volatility dynamics as suggested by the panel VAR estimation.Furthermore, when we augment traditional volatility models with the time series of daily price impact, the augmented models produce significantly more accurate volatility predictions at the one-day ahead forecasting horizon. These volatility predictions also offer economic gains to a mean-variance utility investor in a portfolio setting. JEL code: C53, G14

Financial uncertainty and stock market volatility

“…On the basis of the mixture‐of‐distributions hypothesis, Corsi (2009) develops an additive cascade model of realised volatility—namely, the heterogeneous autoregressive‐realised volatility (HAR‐RV) model, which possesses three heterogeneous volatility components to capture heterogeneous information arrivals. Existing studies further extend the HAR‐RV model to improve its forecasting performance (Andersen et al, 2007; Bollerslev et al, 2018; Corsi et al, 2010; Jiang et al, 2019; Shen et al, 2020; Wang et al, 2016).…”

“…On the basis of the mixture-of-distributions hypothesis, Corsi (2009) develops an additive cascade model of realised volatility-namely, the heterogeneous autoregressiverealised volatility (HAR-RV) model, which possesses three heterogeneous volatility components to capture heterogeneous information arrivals. Existing studies further extend the HAR-RV model to improve its forecasting performance (Andersen et al, 2007;Bollerslev et al, 2018;Corsi et al, 2010;Jiang et al, 2019;Shen et al, 2020;Wang et al, 2016). Motivated by the above strands of the literature, we incorporate the financial uncertainty measure constructed from the unforecastable component of financial time series into long-and short-run volatility components separately to identify the source of volatility determination.…”

Forecasting stock return volatility: Realized volatility‐type or duration‐based estimators

“…They are the heterogeneous autoregressive (HAR) model of Corsi (2009), the heterogeneous autoregressive with continuous volatility and jumps (HAR‐CJ) model of Andersen et al (2007), and the leverage heterogeneous autoregressive with continuous volatility and jumps (LHAR‐CJ) model of Corsi and Reno (2012). The HAR model, exhibiting a multifactor structure to govern the volatility dynamics over different horizons, enjoys empirical success and is widely implemented (see Bollerslev et al, 2016; Jiang et al, 2019; Patton & Sheppard, 2015, among others). Meanwhile, the HAR‐CJ model separates the quadratic variation process into the continuous variation and discontinuous jump component based on the theoretical results of Barndorff‐Nielsen and Shephard (2004).…”