基于Gibbs采样的MIMO信号检测改进算法

针对传统Gibbs采样算法的“失控问题”,提出了一种改进算法,在Gibbs迭代采样过程中加入对噪声方差的估计,消除了失控问题。设计了信号检测流程,将最大似然（ML）检测的搜索过程嵌入Gibbs迭代采样过程,减少了最大似然检测的计算步骤。仿真结果表明,所提算法能估计噪声方差,具有接近最大似然检测的性能。     

A Tutorial on Palm Distributions for Spatial Point Processes

This tutorial provides an introduction to Palm distributions for spatial point processes. Initially, in the context of finite point processes, we give an explicit definition of Palm distributions in terms of their density functions. Then we review Palm distributions in the general case. Finally, we discuss some examples of Palm distributions for specific models and some applications.     

Product Patent Reform and its Impact on Korea's Pharmaceutical Industry

Our analysis focuses on the effect of U.S. government pressure on Korea to adopt product patents for chemical and pharmaceutical products. American pressure began in November 1985 and ended with the Korean Legislature's passage of a new patent law in December 1986. We conduct an event study of the effect of the new patent law on the value of Korean pharmaceutical firms listed on the Korea Stock Exchange. Regression analysis shows that the new law induced excess returns of-74 percent over the 14-month analysis period. The results suggest that adoption of stronger patent laws reduced Korea's wealth. [G14, O34]     

Measuring the cost of liquidity in agricultural futures markets: Conventional and Bayesian approaches

Estimating the cost of liquidity in agricultural futures markets is challenging because bid‐ask spreads are usually not observed. Based on an ability to reflect simulated data from Roll's spread model, we assess the effectiveness of conventional and Bayesian bid‐ask spread estimators under different market conditions. Conventional serial covariance and absolute price change spread estimators appear to be biased. Hasbrouck's Bayesian estimator generates small costs of liquidity whose values depend on the correlation and noise in the data. The absolute value Bayesian estimator is precise and works well under conditions of high levels of noise and correlation usually found in agricultural futures markets. Using data from live cattle (LC) and lean hog (LH) contracts, we find similar patterns of performance that produce economically meaningful cost of liquidity differences.     

Generalized linear mixed models with informative dropouts and missing covariates

Generalized linear mixed models (GLMM) are useful in many longitudinal data analyses. In the presence of informative dropouts and missing covariates, however, standard complete-data methods may not be applicable. In this article, we consider a likelihood method and an approximate method for GLMM with informative dropouts and missing covariates. The methods are implemented by Monte–Carlo EM algorithms combined with Gibbs sampler. The approximate method may lead to inconsistent estimators but is computationally more efficient than the likelihood method. The two methods are evaluated via a simulation study for longitudinal binary data, and appear to perform reasonably well. A dataset on mental distress is analyzed in details.     

Full Likelihood Inference in Normal-Gamma Stochastic Frontier Models

The paper takes up inference in the stochastic frontier model with gamma distributed inefficiency terms, without restricting the gamma distribution to known integer values of its shape parameter (the Erlang form). The paper shows that Gibbs sampling with data augmentation can be used in a computationally efficient way to explore the posterior distribution of the model and conduct inference regarding parameters as well as functions of interest related to technical inefficiency.     

A Stochastic Frontier Analysis of Output Level and Growth in Poland and Western Economies

This paper uses Bayesian stochastic frontier methods to measure the productivity gap between Poland and Western countries that existed before the beginning of the main Polish economic reform. Using data for 20 Western economies, Poland and Yugoslavia (1980–1990) we estimate a translog stochastic frontier and make inference about individual efficiencies. Following the methodology proposed in our earlier work, we also decompose output growth into technical, efficiency and input changes and examine patterns of growth in the period under consideration. This revised version was published online in July 2006 with corrections to the Cover Date.     

On the Application of Markov Chain Monte Carlo Methods to Genetic Analyses on Complex Pedigrees

Markov chain Monte Carlo methods are frequently used in the analyses of genetic data on pedigrees for the estimation of probabilities and likelihoods which cannot be calculated by existing exact methods. In the case of discrete data, the underlying Markov chain may be reducible and care must be taken to ensure that reliable estimates are obtained. Potential reducibility thus has implications for the analysis of the mixed inheritance model, for example, where genetic variation is assumed to be due to one single locus of large effect and many loci each with a small effect. Similarly, reducibility arises in the detection of quantitative trait loci from incomplete discrete marker data. This paper aims to describe the estimation problem in terms of simple discrete genetic models and the single-site Gibbs sampler. Reducibility of the Gibbs sampler is discussed and some current methods for circumventing the problem outlined.     

Comparing risk attitudes of organic and non-organic farmers with a Bayesian random coefficient model

Organic farming is usually considered to be more risky thanconventional farming, but the risk aversion of organic farmerscompared with that of conventional farmers has not been studied.Using a non-structural approach to risk estimation, a Bayesianrandom coefficient model is used to obtain individual Arrow–Prattcoefficients of absolute risk aversion for a sample of Dutchorganic and non-organic arable farmers. The model is estimatedusing Gibbs sampling. The results indicate that organic farmersare significantly less risk averse than their non-organic colleagues.     

Bayesian inference in time series models using kernel quasi likelihoods

The paper takes up Bayesian inference in time series models when essentially nothing is known about the distribution of the dependent variable given past realizations or other covariates. It proposes the use of kernel quasi likelihoods upon which formal inference can be based. Gibbs sampling with data augmentation is used to perform the computations related to numerical Bayesian analysis of the model. The method is illustrated with artificial and real data sets.