Analysis of covariance under inverse Gaussian model

This paper considers the problem of analysis of covariance (ANCOVA) under the assumption of inverse Gaussian distribution for response variable. We develop the essential methodology for estimating the model parameters via maximum likelihood method. The general form of the maximum likelihood estimator is obtained in color closed form. Adjusted treatment effects and adjusted covariate effects are given, too. We also provide the asymptotic distribution of the proposed estimators. A simulation study and a real world application are also performed to illustrate and evaluate the proposed methodology.     

股指期货上市对现货市场的影响——来自中国的实证研究

文章运用EGARCH、多元GARCH、GRANGER非因果性检验、CHOW断点检验等计量经济方法,检验了股指期货上市对股票市场波动的效应、对现货中外联合动态关系的影响、期现之间的先行-滞后关系以及股指期货交易行为对现货波动的影响。研究表明,股指期货的引入加大了现货市场的波动,外生的波动事件通过期货交易,加大了现货市场的波动,引入股指期货后,中外股市的联合动态关系有显著改变。     

Identifying Social Interactions Through Conditional Variance Restrictions

This paper proposes a new method for identifying social interactions using conditional variance restrictions. The method provides a consistent estimate of the social multiplier when social interactions take the “linear‐in‐means” form (Manski (1993)). When social interactions are not of the linear‐in‐means form, the estimator, under certain conditions, continues to form the basis of a consistent test of the no social interactions null with correct large sample size. The methods are illustrated using data from the Tennessee class size reduction experiment Project STAR. The application suggests that differences in peer group quality were an important source of individual‐level variation in the academic achievement of Project STAR kindergarten students.     

The asymptotic covariance matrix of the QMLE in ARMA models

A compact analytical representation of the asymptotic covariance matrix, in terms of model parameters directly, of the quasi maximum likelihood estimator (QMLE) is derived in autoregressive moving average (ARMA) models with possible nonzero means and non-Gaussian error terms. For model parameters excluding the error variance, it is found that the Huber (1967 Huber, P. J. (1967). The behavior of maximum likelihood estimates under nonstandard conditions. Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Volume 1, pp. 221–233. [Google Scholar]) sandwich form for the asymptotic covariance matrix degenerates into the inverse of the associated information matrix. In comparison to the existing result that involves the second moments of some auxiliary variables for the case of zero-mean ARMA models, the analytical asymptotic covariance in this article has an advantage in that it can be conveniently estimated by plugging in the estimated model parameters directly.     

基于协方差阵特征结构的新型控制图研究

本文在深入分析现行广义方差类的统计控制图存在的统计监控缺陷的基础上,提出了对多变量过程的协方差阵的特征向量张成的坐标系的一致性进行统计监控的C控制图,以及对特征值构成的椭球体的相似性进行监控的S控制图,有效地提高了对多变量过程的协方差阵的异常变动的统计监控能力。仿真表明,联合使用C控制图、S控制图及常规均值T²控制图,能够实现对多变量过程的协方差阵与均值的系统监控,显著改善监控效率。     

Linear Models with Doubly Exchangeable Distributed Errors

We study the general linear model (GLM) with doubly exchangeable distributed error for m observed random variables. The doubly exchangeable general linear model (DEGLM) arises when the m-dimensional error vectors are “doubly exchangeable,” jointly normally distributed, which is a much weaker assumption than the independent and identically distributed error vectors as in the case of GLM or classical GLM (CGLM). We estimate the parameters in the model and also find their distributions. We show that the tests of intercept and slope are possible in DEGLM as a particular case using parametric bootstrap as well as multivariate Satterthwaite approximation.     

Estimation of covariance matrix via the sparse Cholesky factor with lasso

In this paper, we discuss a parsimonious approach to estimation of high-dimensional covariance matrices via the modified Cholesky decomposition with lasso. Two different methods are proposed. They are the equi-angular and equi-sparse methods. We use simulation to compare the performance of the proposed methods with others available in the literature, including the sample covariance matrix, the banding method, and the L₁-penalized normal loglikelihood method. We then apply the proposed methods to a portfolio selection problem using 80 series of daily stock returns. To facilitate the use of lasso in high-dimensional time series analysis, we develop the dynamic weighted lasso (DWL) algorithm that extends the LARS-lasso algorithm. In particular, the proposed algorithm can efficiently update the lasso solution as new data become available. It can also add or remove explanatory variables. The entire solution path of the L₁-penalized normal loglikelihood method is also constructed.     

Public Response to the Tokai Nuclear Accident

This article discusses the influence of the September 30, 1999 nuclear accident in Tokai village (Japan) on the public's attitudes toward nuclear power in Japan. The data used in this report were taken from the results of two surveys conducted mainly to measure the attitudes of the Japanese public with regard to the use of nuclear power in Japan. The first survey was done before the accident in District 23 in Tokyo and also in Osaka and Nagoya. The second survey, which took place after the accident in District 23 in Tokyo and in Osaka and Nagoya, also included residents in a number of other cities of various sizes throughout Japan. The results of the two surveys showed that (1) acceptability of and trust in nuclear power operation had decreased, (2) perceived accident likelihood and public interest had significantly increased, and (3) there had been neither significant nor even a small change in the public's self-rated knowledge about nuclear power or their distrust of the government. The results also showed that the ratio of nuclear power generation opponents to total respondents had considerably increased (7% to 23%) whereas nuclear power generation supporters had moderately decreased (1% to 12%).     

Standard errors for EM estimation

The EM algorithm is a popular method for computing maximum likelihood estimates. One of its drawbacks is that it does not produce standard errors as a by-product. We consider obtaining standard errors by numerical differentiation. Two approaches are considered. The first differentiates the Fisher score vector to yield the Hessian of the log-likelihood. The second differentiates the EM operator and uses an identity that relates its derivative to the Hessian of the log-likelihood. The well-known SEM algorithm uses the second approach. We consider three additional algorithms: one that uses the first approach and two that use the second. We evaluate the complexity and precision of these three and the SEM in algorithm seven examples. The first is a single-parameter example used to give insight. The others are three examples in each of two areas of EM application: Poisson mixture models and the estimation of covariance from incomplete data. The examples show that there are algorithms that are much simpler and more accurate than the SEM algorithm. Hopefully their simplicity will increase the availability of standard error estimates in EM applications. It is shown that, as previously conjectured, a symmetry diagnostic can accurately estimate errors arising from numerical differentiation. Some issues related to the speed of the EM algorithm and algorithms that differentiate the EM operator are identified.