信用风险模型的贝叶斯改进研究

基于小样本数据和外部先验信息,本文运用贝叶斯(Bayes)估计量来改进信用风险模型的违约预测力。同时,运用中国上市公司财务数据,分别对贝叶斯估计量和标准Logit估计量进行了模拟估计,并通过统计量AUC值和布莱尔分数(Brier Score)对其预测精度进行比较。结果表明,贝叶斯估计量具有更高的预测精度和稳定性。     

Market Reform and Efficiency: The Case of KOSPI200 Options

The Korean government and exchange have identified a need to regulate excessive speculative trading and to protect domestic individual investors from foreign and professional traders. As such, they have proposed an options market reform that requires higher levels of margin accounts for options trading and that increases the basic options multipliers in the KOSPI200 options market. This study examines how this market reform affects the price disagreement and adjustment behaviors of the index options market. Our analyses indicate that the efficiency and information quality of out-of-the-money options trades have increased since the reform took effect.     

Seasonals,price adjustment delays and estimated systematic risk

This article analyzed potential interactions between seasonals and price adjustment delays on estimated systematic risk. It was shown that seasonals in unobservable true security returns can induce inconsistencies into the generalized Scholes and Williams estimator of systematic risk. An alternative estimator was proposed that is consistent in the presence of seasonals in the unobservable true returns. The direction of induced bias is unpredictable a priori, thereby representing a potentially important research consideration in market efficiency tests using abnormal returns. NASDAQ and Dow Jones 30 Industrial return data for the period 1983–87 were used to evaluate the proposed estimator against the OLS and generalized Scholes and Williams (GSW) alternatives. The absolute difference between the GSW and our estimator, that is the seasonal-induced bias, for NASDAQ stocks was negatively correlated with market capitalization. Moreover, seasonal-induced bias was larger for NASDAQ stocks than more highly capitalized Dow stocks. These empirical findings indicate that seasonals and price adjustment delays can interact to bias estimated systematic risk, where price adjustment delays would be projected to be more acute for smaller capitalization stocks.     

Asymmetric Impact of Oil Price Shock on Stock Market in China: A Combination Analysis Based on SVAR Model and NARDL Model

This article integrates the SVAR model and nonlinear ARDL (NARDL) model to analyze the long-run and short-run asymmetric effect of structural oil price shocks on the Chinese stock market. We reveal that the demand-side shocks of oil price have a significant impact on the Chinese stock market in both short and long run, but the supply shock is an exception. In terms of asymmetric nature, there is no evidence of asymmetric impact when it refers to the supply shock and the oil-specific demand shock on stock market, and only the aggregate demand shock has asymmetric effect in short run.     

Administrative division adjustments and stock price comovement: Evidence from the city–county mergers in China

Under the background of Chinese market segmentation, whether government-led administrative division adjustments can promote regional economic integration is a practical issue. Taking interregional firms’ stock price comovement as a micro measurement of regional integration, this paper investigates the regional integration effect of administrative division adjustments, i.e., city–county mergers. We find that stock price comovement between county-level and municipal district-level firms in the merged counties and municipal districts significantly improve after city–county mergers, particularly in regions with a higher degree of market segmentation and lower degree of marketization. We further find that the increase in stock price comovement caused by city–county mergers emerges from the increase in comovement of real activities between firms in the merged counties and municipal districts. Taken together, our results suggest that government-led administrative division adjustments effectively promote regional integration.     

A Bayesian Approach to the Construction and Comparison of Alternative House Price Indices

Several repeat-sales models have been advanced over the years for estimating real estate price indices. This article proposes a general model which incorporates earlier works as special cases and compares the alternative repeat-sales models using posterior odds ratios as criteria. While the existing literature estimates the real estate indices from the sampling point of view, in this article indices are constructed and then compared using a Bayesian approach. In general, the two-error term models outperform the one-error models. The model with a nontemporal component proposed by Goetzmann and Spiegel is found to be superior in three out of four cities. There is a significant discrepancy among the returns and indices obtained from different models.     

International financial integration through the law of one price: The role of liquidity and capital controls

This paper takes advantage of the fact that some stocks trade both in domestic and international markets to characterize the degree of international financial integration. The paper argues that the cross-market premium (the ratio between the domestic and the international market price of cross-listed stocks) provides a valuable measure of international financial integration and the effectiveness of capital controls. Using autoregressive (AR) models to estimate convergence speeds and non-linear threshold autoregressive (TAR) models to identify non-arbitrage bands, the paper shows that price deviations across markets are rapidly arbitraged away and bands are narrow, particularly so for liquid stocks. The paper also shows that regulations on cross-border capital flows effectively segment domestic markets. As expected, the effects of both types of capital controls are asymmetric but in the opposite direction: controls on outflows induce positive premia, while controls on inflows generate negative premia. Both vary with the intensity of capital controls.     

Price Limits and Beta

Price limits, which restrict daily price changes of a stock within a pre-specified range, make the stochastic properties of observed returns deviate from those of true returns, and hence lead to a biased estimates of the market model parameters. To investigate the impacts of price limits on the market model parameters, especially on beta, the restricted regression analysis is performed as well as the two-pass regression analysis used in examining the intervalling effect bias on beta. Empirical results suggest that, when prices are observed within a pre-specified bound, the estimates of beta using ordinary least squares substantially understate the true beta and suffer more from the intervalling effect bias. However, the delay effect of price limits on the adjustment of a security's price does not last too long, that is, remaining information is reflected on the subsequent day's stock prices very rapidly.