Concurrent processing of heteroskedastic vector-valued mixture density models

We introduce a combined two-stage least-squares (2SLS)–expectation maximization (EM) algorithm for estimating vector-valued autoregressive conditional heteroskedasticity models with standardized errors generated by Gaussian mixtures. The procedure incorporates the identification of the parametric settings as well as the estimation of the model parameters. Our approach does not require a priori knowledge of the Gaussian densities. The parametric settings of the 2SLS_EM algorithm are determined by the genetic hybrid algorithm (GHA). We test the GHA-driven 2SLS_EM algorithm on some simulated cases and on international asset pricing data. The statistical properties of the estimated models and the derived mixture densities indicate good performance of the algorithm. We conduct tests on a massively parallel processor supercomputer to cope with situations involving numerous mixtures. We show that the algorithm is scalable.     

中国经济周期条件波动性特征的统计分析

在ARCH族模型动态地刻画了中国经济周期的条件波动性特征的基础上,分析了有关经济变量的波动机制、波动率对经济变量条件均值的影响程度与方向以及经济变量在扩张阶段和紧缩阶段波动力度差异等问题。得出以下主要结论：中国经济周期波动具有条件异方差性、持续性和非对称性的特征;经济的稳定有利于经济产出的持续增长;财政政策具有稳定经济、促进经济产出增长率提高的效应,而且这种效应具有一定的持续性与滞后性;国外需求受国内经济影响较小,具有其自身的发展规律。     

基于GMDH组合的中国GDP预测模型研究

文章对中国季度GDP分别建立了ARIMA和ARCH模型,并利用GMDH自组织建模方法提出了新的组合预测模型。模型预测结果及对比表明,基于GMDH组合的GDP预测模型的拟合和预测效果,在经济正常增长或出现较大波动时都具有较高的可靠性与准确性。文章还使用Bon-ferroni-Dunn检验方法进一步验证了组合模型的拟合能力要优于单一模型。     

Testing for multivariate heteroscedasticity

In this article, we propose a testing technique for multivariate heteroscedasticity, which is expressed as a test of linear restrictions in a multivariate regression model. Four test statistics with known asymptotical null distributions are suggested, namely the Wald, Lagrange multiplier (LM), likelihood ratio (LR) and the multivariate Rao F-test. The critical values for the statistics are determined by their asymptotic null distributions, but bootstrapped critical values are also used. The size, power and robustness of the tests are examined in a Monte Carlo experiment. Our main finding is that all the tests limit their nominal sizes asymptotically, but some of them have superior small sample properties. These are the F, LM and bootstrapped versions of Wald and LR tests.     

A pseudo restricted MLE under multivariate order restrictions and its algorithm

Abstract

It is shown in this paper that a quasi order for the vectors in R^p is a cone induced if and only if the order is preservable under limits and under linear combinations with non-negative coefficients. For the mean vectors in MANOVA subject to the restriction of simple ordering, a pseudo restricted MLE is proposed. This estimator is a matrix projection onto a closed convex set inside the restricted domain. An algorithm for the pseudo restricted MLE is developed, that computes the matrix projections using only vector projections.     

New Product Development Decision‐Making Effectiveness: Comparing Individuals,Face‐To‐Face Teams,and Virtual Teams*

A total of 411 subjects participated in two decision‐making experiments in order to examine the effectiveness of new product development project continuation decisions. Using escalation of commitment theory, in Study 1, individual versus face‐to‐face team decision‐making effectiveness was compared. Study 2, an extension of Study 1, compared the new product development decision‐making effectiveness of individuals, face‐to‐face teams, and virtual teams. A virtual team is a geographically and temporally dispersed and electronically communicating work group. In Study 2, the virtual teams communicated asynchronously via groupware technology. Our findings suggest that teams make more effective decisions than individuals, and virtual teams make the most effective decisions.     

Adaptive testing in arch models

Specification tests for conditional heteroskedasticity that are derived under the assumption that the density of the innovation is Gaussian may not be powerful in light of the recent empirical results that the density is not Gaussian. We obtain specification tests for conditional heteroskedasticity under the assumption that the innovation density is a member of a general family of densities. Our test statistics maximize asymptotic local power and weighted average power criteria for the general family of densities. We establish both first-order and second-order theory for our procedures. Simulations indicate that asymptotic power gains are achievable in finite samples.     

上证国债指数“杠杆效应”的检验与分析

上证国债指数是中国国债价格的一个很好标尺。研究国债指数的波动性特征对于认识国债市场的波动有着很大的帮助。对近三年来上证国债指数进行实证分析，指出了其收益率分布的“尖峰厚尾”以及条件异方差的存在。同时采用三种不对称ARCH模型，TARCH，EGARCH，非对称CARCH验证了该指数的不对称“杠杆效应”。从行为金融学和市场参与者交易行为的角度来分析国债指数为什么存在“杠杆效应”的原因。