Robustness issues in multilevel regression analysis

A multilevel problem concerns a population with a hierarchical structure. A sample from such a population can be described as a multistage sample. First, a sample of higher level units is drawn (e.g. schools or organizations), and next a sample of the sub‐units from the available units (e.g. pupils in schools or employees in organizations). In such samples, the individual observations are in general not completely independent. Multilevel analysis software accounts for this dependence and in recent years these programs have been widely accepted. Two problems that occur in the practice of multilevel modeling will be discussed. The first problem is the choice of the sample sizes at the different levels. What are sufficient sample sizes for accurate estimation? The second problem is the normality assumption of the level‐2 error distribution. When one wants to conduct tests of significance, the errors need to be normally distributed. What happens when this is not the case? In this paper, simulation studies are used to answer both questions. With respect to the first question, the results show that a small sample size at level two (meaning a sample of 50 or less) leads to biased estimates of the second‐level standard errors. The answer to the second question is that only the standard errors for the random effects at the second level are highly inaccurate if the distributional assumptions concerning the level‐2 errors are not fulfilled. Robust standard errors turn out to be more reliable than the asymptotic standard errors based on maximum likelihood.     

Pseudolikelihood estimation of the stochastic frontier model

Stochastic frontier analysis is a popular tool to assess firm performance. Almost universally it has been applied using maximum likelihood (ML) estimation. An alternative approach, pseudolikelihood (PL) estimation, which decouples estimation of the error component structure and the production frontier, has been adopted in both the non-parametric and panel data settings. To date, no formal comparison has yet to be conducted comparing these methods in a standard, parametric cross-sectional framework. We produce a comparison of these two competing methods using Monte Carlo simulations. Our results indicate that PL estimation enjoys almost identical performance to ML estimation across a range of scenarios and performance metrics, and for certain metrics, outperforms ML estimation when the distribution of inefficiency is incorrectly specified.     

上海股市收益与波动的周内效应研究

股市周内效应一直是金融投资者关注的焦点问题，许多学者已做了大量研究，但多数文献将收益与波动的周内效应分开来进行研究和检验，忽视了波动与收益的共生性，其结果缺乏严密性和说服力。针对这种情况，提出平行数据GARCH模型并给出了参数的极大似然估计方法，进而对上海股市收益和波动的周内效应进行检验，既反映收益与风险存在共生关系，又避免了分别判断收益和波动的周内效应所致的缺点。     

Thailand-Japan＇s Volume of Trade Influencing on Socio-economic and Well-Being Development Using Maximum Entropy Bootstrap

This paper is preliminary information on some empirical findings based on an analysis of Thailand＇s macroeconomic data. The purpose of the study is to quantify social-economic well-being development with special emphasis on poverty. The maximum entropy bootstrap was presented that did not require the property of stationarity. Moreover, the methodology was stratified both the ergodic theorem and the central limit theorem. Thailand＇s GINI index and Thailand-Japan＇s volume of trade data cover the 14-year period from 1998 to 2011. There is a statistically significant negative non-linear relationship between endogenous Thailand＇s GINI index and exogenous Thailand-Japan＇s volume of trade. The results confirmed that every lagged one percent increase in Thailand-Japan＇s volume of trade influenced by a decrease of Thailand＇s GINI index coefficient 0.039 percent with the asymmetric around 0-almost closed to range condition -0.034824348 at 2.5%.     

Gravity Models of Interregional Migration in Indonesia

This article explores the determinants of interregional migration in Indonesia. Employing basic and modified (extended) gravity models, and using data from the 2000 and 2010 Population Censuses and the 2005 Intercensal Population Survey, we test Long’s (1985) hypothesis that in the early stages of population redistribution, economic development is positively related to a concentration of the population. Using per-capita GDP as a proxy for income and as an indicator of economic development, we find that migration in Indonesia is indeed directed towards more developed regions. This finding supports the notion that regional disparities in development are an important factor in interregional migration in Indonesia. In line with classical gravity models, our findings show that distance is negatively related to the size of migration flows. However, unlike previous studies of interprovincial migration in Indonesia, we find that the effect of distance has weakened over time.     

Fear of Floating and Pegging: A Simultaneous Choice Model of De Jure and De Facto Exchange Rate Regimes in Developing Countries

We present an analysis of the determinants of de jure and de facto exchange rate regimes based on a panel probit model with simultaneous equations. The model is estimated using simulation-based maximum likelihood methods. The empirical results suggest a triangular structure of the model such that the choice of de facto regimes depends on the choice of de jure regimes but not vice versa. This gives rise to a novel interpretation of regime discrepancies.

广义空间调制系统的正则化OMP检测算法

在广义空间调制（GSM）系统中,最大似然（ML）检测可以取得最优的检测性能,然而其计算复杂度随激活天线数的增加急剧增长。针对这一问题,提出了一种基于稀疏重构理论的低复杂度检测算法——正则化正交匹配追踪（ROMP）算法。该算法首先根据信道矩阵和当前残差的内积选取多个候选激活天线索引,接着对候选天线索引按正则化标准进行可靠性验证,剔除错误索引,缩小信号的搜索空间,最后通过求解最小二乘问题估计信号。仿真结果表明,与经典的正交匹配追踪（OMP）算法相比,所提算法以少许复杂度的增加为代价极大提升了检测性能,能够在检测性能与复杂度之间取得更好的折中。     

Small-sample Properties of Panel Spatial Autoregressive Models: Comparison of the Bayesian and Maximum Likelihood Methods

Abstract

This article considers autoregressive (SAR) models. We method to estimate the parameters of likelihood (ML) method. Our Bayesian by the Monte Carlo studies. We found the efficient as the ML estimators.