基于金融高频数据的ACD模型非参数设定检验

 目前关于ACD的实证研究已经十分丰富，却很少有人把注意力放在ACD及其扩展模型设定的检验上，本文采用的D检验就是通过衡量残差密度函数的参数和非参数估计值之间的紧密程度，来检验模型设定的优劣。     

A generalized-moments specification test for the logistic link

The authors consider the problem of testing the validity of the logistic regression model using a random sample. Given the values of the response variable, they observe that the sample actually consists of two independent subsets of observations whose density ratio has a known parametric form when the model is true. They are thus led to propose a generalized-moments specification test in detail. In addition, they show that this test can be derived using Neyman's smooth tests for goodness of fit. They present simulation results and apply the methodology to the analysis of two real data sets.     

Robust Bayesian bonus-malus premiums under the conditional specification model

The conditional specification technique introduced by Arnold et al. (Conditional specification of statistical models. Springer series in statistics. Springer, New York, 1999) was used in Sarabia et al. (Astin Bull 34(1):85–98, 2004) to obtain bonus-malus premiums. The Poisson distribution for which the parameter is a function of the classical structure parameter was used and a new class of prior distributions appeared in a natural way. This model contains, as a particular case, the classical compound Poisson model. In the present paper, the Bayesian robustness of this new model is examined and found to be much more robust than in the classical model in Gómez et al. (Insur Math Econ 31:105–113, 2002). For the present study, the moment conditions on the prior distribution are required. Examples, with real data, are given to illustrate our ideas under the net and exponential premium principles.     

检验式设定错误对时间序列单位根检验小样本性质的影响

首先对单位根检验的两类常见的数据生成系统进行比较,然后利用蒙特卡洛实验研究了时间序列单位根检验式的设定问题。研究发现在利用DF检验和DF-GLS检验进行时间序列的单位根检验时,检验式设定错误直接影响着检验结果,尤其在推断时间序列是趋势平稳过程还是有时间趋势项的随机游走过程或有二阶时间趋势多项式的随机游走过程时,检验式的错误设定很容易将趋势平稳过程误判为非平稳过程。     

Combining Robust and Traditional Least Squares Methods: A Critical Evaluation

The combination of robust and least squares procedures has frequently been recommended as a useful strategy for constructing models. The application of this strategy to a real-world data set resulted in a model with an incorrect functional form. Additional in-depth investigations into the nature of the application, combined with data-error corrections, made possible the construction of a satisfactory model. The results of the modeling activity were evaluated in terms of model face-validity, the predictive performance on a holdout data set, and the ability to meet user requirements. The findings of this study demonstrate the danger of model-form misspecification when one mistakenly assumes that the combination of robust and least squares procedures compensates for a lack of knowledge about the processes underlying the generation of the data.     

Bayesian estimation of incomplete data using conditionally specified priors

In this article, a class of conjugate prior for estimating incomplete count data based on a broad class of conjugate prior distributions is presented. The new class of prior distributions arises from a conditional perspective, making use of the conditional specification methodology and can be considered as the generalization of the form of prior distributions that have been used previously in the estimation of incomplete count data well. Finally, some examples of simulated and real data are given.     

Tables of confidence limits on the tail area of the normal distribution

Tables are given of confidence limits on tail areas, γ, of the normal distribution, where γ = P{Y ≥ L}, and where L is a given number, and Y is normally distributed with unknown mean, μ, and unknown variance, σ².     

On discrete weighted distributions and their use in model choice for observed data

This paper provides a brief structural perspective of discrete weighted distributions in theory and practice.. It develops a unified view of previous work involving univariate and bivariate models with some new results pertaining to mixtures, form-invariance and Bayesian inference     

Market channel selections by US nursery plant producers: a multivariate nonparametric fractional regression analysis

Availability of market channel alternatives has helped the growth of ornamental plant sales in the United States. To identify the factors affecting the choice and allocation of outputs to different market channels by nursery producers, we first use a mixture of experts model to select clusters of homogenous subpopulations of US nursery producers based on a 2009 National Nursery Survey. The impact of growers’ business characteristics on shares of sales to these channels was estimated using multivariate parametric and nonparametric fractional regression models. Specification tests indicated a nonparametric model was superior to a parametric model in some clusters. Important explanatory variables affecting the sales volume to different channels were sales of plant groups, kinds of contract sales, promotional expenses, and farm size. Results indicated the existence of clear market segmentation of nursery producers in the United States.