Junji Shimada

Estimation of Stochastic Volatility Models: An Approximation to the Nonlinear State Space Representation

ABSTRACT The stochastic volatility (SV) model can be regarded as a nonlinear state space model. This article proposes the Laplace approximation method to the nonlinear state space representation and applies it for estimating the SV models. We examine how the approximation works by simulations as well as various empirical studies. The Monte Carlo experiments for the standard SV model indicate that our method is comparable to the Monte-Calro Likelihood (MCL; Durbin and Koopman, 1997), Maximum Likelihood (Fridman and Harris, 1998), and Markov chain Monte Carlo (MCMC) methods in the sense of mean square error in finite sample. The empirical studies for stock markets reveal that our method provides very similar estimates of coefficients to those of the MCL. We show a relationship of our Laplace approximation method to importance sampling.

DYNAMIC WELFARE COST OF STAGNATION: AN ALTERNATIVE MEASURE TO THE LUCAS–OBSTFELD MODEL

The present paper proposes an alternative measure to the Lucas–Obstfeld model to analyze the welfare costs of stagnation, and provides a practical illustration of both the Lucas–Obstfeld model and the alternative model. Compared with the Lucas–Obstfeld model, the alternative model can evaluate: (i) whether policy was implemented in a timely fashion; (ii) whether the policy cost was expensive compared with the cost of stagnation; and (iii) whether the policy implemented was effective or whether an additional policy is required.

The Impacts Of The Imf-Supported Structural Reform Program On Asian Stock Market Efficiency

We investigate the impact of the IMF-supported structural reform program in the 1997 Asian crisis on stock market efficiency using the before–after, with–without and event study approaches by applying a time-varying parameter model to eight Asian stock markets. All the supported countries, including Indonesia and Korea, but not Thailand, experienced significantly improved market efficiency after the implementation of the program, implying a positive effect of the program according to the before–after approach. Among the nonsupported countries, China, Taiwan and Malaysia did not improve efficiency (however, Hong Kong and Singapore did) after the start of the crisis, providing some evidence of a positive effect according to the with–without approach. The Thailand, Indonesia and Korean markets showed positive abnormal returns on the days or days following policy announcements in this IMF-supported program, indicating positive effects of the policy according to the event study approach. These findings suggest that the IMF program was successful during the 1997 Asian financial crisis and that it was helpful in resolving the recent global financial crisis.

ASYMMETRIC INTERNATIONAL TRANSMISSION IN THE CONDITIONAL MEAN AND VOLATILITY TO THE JAPANESE MARKET FROM THE US: EGARCH VERSUS SV MODELS

This paper investigates whether the upturns and downturns of the US market exert asymmetric influence on the conditional mean and volatility of the Japanese market using the daily returns on stock price indices. Using both the EGARCH and SV models, which simultaneously allow two kinds of asymmetric international transmissions across the markets, the result reconfirms the symmetric transmission in the conditional mean obtained by Bahng and Shin (2003) and the asymmetric transmission in the conditional volatility obtained by Koutmos and Booth (1995) although each of them analyzed only one spillover effect separately. Although the EGARCH and SV models lead to similar results for the spillover effects, the SV model is preferred to the EGARCH model based on Lagrange Multiplier test for the hypothesis of the EGARCH against the SV. The shock to volatility in the US market with the SV model is asymmetrically transmitted to the volatility in the Japanese market.