Penalized likelihood smoothing in robust state space models

In likelihood-based approaches to robustify state space models, Gaussian error distributions are replaced by non-normal alternatives with heavier tails. Robustified observation models are appropriate for time series with additive outliers, while state or transition equations with heavy-tailed error distributions lead to filters and smoothers that can cope with structural changes in trend or slope caused by innovations outliers. As a consequence, however, conditional filtering and smoothing densities become analytically intractable. Various attempts have been made to deal with this problem, reaching from approximate conditional mean type estimation to fully Bayesian analysis using MCMC simulation. In this article we consider penalized likelihood smoothers, this means estimators which maximize penalized likelihoods or, equivalently, posterior densities. Filtering and smoothing for additive and innovations outlier models can be carried out by computationally efficient Fisher scoring steps or iterative Kalman-type filters. Special emphasis is on the Student family, for which EM-type algorithms to estimate unknown hyperparameters are developed. Operational behaviour is illustrated by simulation experiments and by real data applications. Received: March 1998     

Gibson土体非线性大变形固结方程的解析解

探讨了土体一维非线性大变形固结问题的解析解法,给出了土体在单面排水和双面排水条件下、以孔隙比为变量的G ibson土体一维非线性大变形固结简化控制方程的解析解.利用该解析解分析了土体一维非线性大变形固结的性状,将土体一维非线性大变形固结理论解与Terzagh i一维小变形固结理论解进行了比较.研究结果表明,该解析解可用于土体一维非线性大变形固结问题其他数值解的验证.     

Optimal Redistribution with Unobservable Preferences for an Observable Merit Good

This paper considers a government thatseeks both to redistribute income and to encourage or discouragethe consumption of a certain good. This good is assumed to beeither a merit or demerit good. Individuals differ in their exogenousincome and in their preferences for the merit good. The onlyvariable the government can perfectly observe is each individual'sconsumption of the merit good. In order to account for meritgood considerations, we consider a modification of the utilitariansocial welfare function in which the government imposes uniformpreferences, despite the heterogeneous individual preferences,at a level which will depend on the merit or demerit nature ofthe observable good. We derive the optimal nonlinear redistributivepolicy and compare our results to the ones that would be obtainedunder a utilitarian social welfare function that respects theown preferences of individuals.     

Negotiating Complex Contracts

Work to date on computational models of negotiation has focused almost exclusively on defining contracts consisting of one or a few independent issues and tractable contract spaces. Many real-world contracts, by contrast, are much more complex, consisting of multiple inter-dependent issues and intractably large contract spaces. This paper describes a simulated annealing based approach appropriate for negotiating such complex contracts that achieves near-optimal social welfares for negotiations with binary issue dependencies.     

The relevance of business cycles in forecasting international tourist arrivals

Casual empiricism suggests that there may be a cyclical trend associated with international tourist arrivals in which variation around the linear trend can be formed by the interaction with other cyclical phenomena. This paper employs a simple model that incorporates a linear trend and sine function to capture these two characteristics in forecasting international tourist arrivals in Hong Kong. The model is extended to include a set of sine functions through the application of Fourier analysis to account for situations in which more than one phenomenon may be present in the time series. The forecasting accuracy of the model is compared with other forecasting approaches. Evaluation of the results using the mean absolute percentage error measure show that the forecasting performance of the extended model with a linear trend and two sine functions is superior in terms of accuracy when compared with other forecasting models.     

Oil price shocks and producer prices in Taiwan: an application of non-linear error-correction models

This study estimated the short-term and long-term pass-through effects of oil prices on inflation in Taiwan from 1981M1-2011M5, employing the producer price general index and various basic sub-indices for evaluation. The empirical results show that oil prices have long-term and short-term pass-through effects on Taiwan’s producer price indices. Moreover, producer prices have significant non-linear error-correction relationships with the oil price, output and wages, suggesting asymmetric and time-variant properties of error correction. When the deviation of price in the equilibrium is greater, the error-correction adjustment will be faster. Our findings could therefore enable the monetary authorities and manufacturers to formulate a more effective policy from the oil price shocks.     

公司创业的创新环境比较

以企业家创业(entrepreneurship)相对应的近期来研究热点公司创业(corporate entrepreneurship)为分析核心,着重研究了技术创新、企业创新和管理创新是产生公司创业的创新环境。并针对这些研究热点表明了本文的一些观点。如技术创新的衡量指标、引导创新的非线性思维和企业创新的推动作用等,是作者在创新与增长领域长期研究的部分观点。     

Forward-looking portfolio selection with multivariate non-Gaussian models

In this study, we suggest a portfolio selection framework based on time series of stock log-returns, option-implied information, and multivariate non-Gaussian processes. We empirically assess a multivariate extension of the normal tempered stable (NTS) model and of the generalized hyperbolic (GH) one by implementing an estimation method that simultaneously calibrates the multivariate time series of log-returns and, for each margin, the univariate observed one-month implied volatility smile. To extract option-implied information, the connection between the historical measure P and the risk-neutral measure Q, needed to price options, is provided by the multivariate Esscher transform. The method is applied to fit a 50-dimensional series of stock returns, to evaluate widely known portfolio risk measures and to perform a forward-looking portfolio selection analysis. The proposed models are able to produce asymmetries, heavy tails, both linear and non-linear dependence and, to calibrate them, there is no need for liquid multivariate derivative quotes.     

Investor sentiment and the prediction of stock returns: a quantile regression approach

We employ quantile regression to provide a detailed picture of the stock return forecasting ability of investor sentiment. We find that investor sentiment predicts aggregate stock returns at lower quantiles. However, the forecasting power is lost at upper quantiles. The results are robust after controlling for a comprehensive set of macroeconomic and financial predictors and for characteristic portfolios. We also show that investor sentiment consists mainly of cash flow news and contains little information about discount rate news. The ability to forecast cash flows increases gradually from the lower quantiles to upper quantiles. Our results do not support that the ability of investor sentiment to predict stock returns comes from a rational forecast of future cash flows.