Estimating Behavioural Heterogeneity Under Regime Switching

Abstract

Financial markets are typically characterized by high (low) price level and low (high) volatility during boom (bust) periods, suggesting that price and volatility tend to move together with different market conditions/states. By proposing a simple heterogeneous agent model of fundamentalists and chartists with Markov chain regime-dependent expectations and applying S&P 500 data from January 2000 to June 2010, we show that the estimation of the model matches well with the boom and bust periods in the US stock market. In addition, we find evidence of time-varying behavioural heterogeneity within-group and that the model exhibits good forecasting accuracy. * Acknowledgement: This work was initiated while Huanhuan Zheng was visiting the School of Finance and Economics at the University of Technology, Sydney (UTS), whose hospitality she gratefully acknowledges. Financial support for Chiarella and He from the Australian Research Council (ARC) under Discovery Grant (DP0773776) is gratefully acknowledged. We would like to thank Willi Semmler and Lucas Bernard, the editors of this special issue, Remco Zwinkels and participants of the 2010 Guangzhou Conference on Nonlinear Economic Dynamics and Financial Market Modelling for helpful comments. The usual caveat applies.

Structural estimation of behavioral heterogeneity

“…(i) The two strategies are used commonly in practice (Allen & Taylor, 1990). (ii) Models accounting for two such strategies are powerful in explaining financial market phenomena such as bubbles and crashes (Huang, Zheng, & Chia, 2010;Lux, 1995) and providing empirical specifications that outperform random walk (Chiarella et al 2012). (iii) Due to resource constraints, it is reasonable to prioritize investment strategies with good tracking records, supported by theoretical or empirical foundations; it is costly to hire a large number of financial advisors to conduct various analysis.…”

“…In terms of modeling, the dynamics in trading heterogeneity has been modeled by latent boom-bust market states (Chiarella, He, Huang, & Zheng, 2012), real business cycles (Lof, 2012), and switching stochastic processes (Brock & Hommes, 1998). In terms of modeling, the dynamics in trading heterogeneity has been modeled by latent boom-bust market states (Chiarella, He, Huang, & Zheng, 2012), real business cycles (Lof, 2012), and switching stochastic processes (Brock & Hommes, 1998).…”

“…In terms of estimation methods, existing empirical works of HAM mostly use nonlinear least squares (Boswijk, Hommes, & Manzan, 2007;Chiarella et al 2012;Frijns et al 2010), except that Franke and Westerhoff (2012) utilize the simulated method of moments (SMM). We derive an explicit formula of the pricing mechanism that implies moment restrictions in closed form, which simplifies and speeds up the estimation.…”

“…Our paper makes several contributions to the literature. In terms of modeling, the dynamics in trading heterogeneity has been modeled by latent boom-bust market states (Chiarella, He, Huang, & Zheng, 2012), real business cycles (Lof, 2012), and switching stochastic processes (Brock & Hommes, 1998). In particular, Markov transition of discrete regimes is popular in modeling the switching processes, and finds many empirical applications in the stock market, commodity market, and derivative market (Eichholtz, Huisman, & Zwinkels, 2015;Frijns, Lehnert, & Zwinkels, 2010;Jongen, Verschoor, Wolff, & Zwinkels, 2012;Ter Ellen, Verschoor, & Zwinkels, 2013).…”

Essays on international financial integration

“…In fact, Ang and Bekaert (2002) postulated that the correlations between international equity market returns tend to increase during highly volatile bear market and showed that such process can be well captured by a multiple regime switching process. Chiarella et al (2012) found that the price evolution of S&P 500 can be characterized by a two-state Markov regime switching HAM model which leads to improved forecast accuracy. In light of these findings, we will assess our model using nonlinear techniques by allowing for two-state regime switching.…”