Empirical likelihood inference for a common mean in the presence of heteroscedasticity

The authors develop empirical likelihood (EL) based methods of inference for a common mean using data from several independent but nonhomogeneous populations. For point estimation, they propose a maximum empirical likelihood (MEL) estimator and show that it is n‐consistent and asymptotically optimal. For confidence intervals, they consider two EL based methods and show that both intervals have approximately correct coverage probabilities under large samples. Finite‐sample performances of the MEL estimator and the EL based confidence intervals are evaluated through a simulation study. The results indicate that overall the MEL estimator and the weighted EL confidence interval are superior alternatives to the existing methods.     

Family income at the bottom and at the top: Income sources and family characteristics

Attention has recently been focused on wealth as a source of long-term economic security and on wealth ownership as a crucial aspect of the racial economic divisions in the United States. This literature, however has been concerned primarily with the wealth gap between poor and middle-class families, and between the white and black middle class. In this paper, we investigate the incomes of families at the top and bottom of the family income distribution. We examine the sources of income and the demographic characteristics of these high-income and low-income families using family level data from the 1988 to 2003 Current Population Surveys.     

OCCUPATIONAL DISPARITY IN A MIGRANT METROPOLIS: A CASE STUDY OF ATLANTA

ABSTRACT

As the populations of immigrant and migrant groups swell in Atlanta, Georgia, the nature of race and ethnic relations in the metropolis, one historically dominated by a black-white dichotomy, is changing. In the 2000 census nearly 40% of the Atlanta metropolitan area's working-age population was an immigrant, a return migrant, or a northern-born migrant. We use the 5% Public Use Microdata Sample (PUMS) from the 2000 census to examine over- and under-occupational representation in the metropolitan area among migrant and nonmigrant groups. The results indicate that there remains a racial and ethnic hierarchy that is related to occupational queues and segmented labor markets. By comparing different types of migrant groups, and by taking into account race, ethnicity, and nativity, we more clearly delineate economic stratification in the metropolitan region.     

BaernsteinⅡ空间的几何特征

本文给出了BaernsteinⅡ空间X_B的单位球U(X_B)的极端点和非极端点的判别法(充要条件和充分条件),并举一些例子。作为推论,X_B不是严格凸的。     

Deep South demography: New immigrants and racial hierarchies

ABSTRACT

This research documents the racial-ethnic hierarchy in six Deep South states by focusing on native- and foreign-born groups and key socioeconomic outcomes including labor force participation, employment status, and poverty. The proportion of immigrants from Latin American countries entering these states has increased dramatically over the last 20 years; this is true to a lesser extent for immigrants from Asian nations. Using Public Use Microdata Samples and American Community Survey data, we find that although whites remain dominant demographically and economically, native- and foreign-born Asians are beginning to share in that economic position. Results also indicate that African Americans and Hispanic immigrants are often at the bottom of the hierarchy, but that the order of the groups can shift depending on the context and measure. Overall, the white-black framework of race relations may be slowly transforming into a complicated, multifaceted racial hierarchy but blacks still remain considerably disadvantaged in the Deep South.     

Improved confidence estimators for Fieller's confidence sets

Fieller's confidence set C_F for ratios of location parameters, although of great importance in practice, is often cited as an example to criticize frequentist theory. The reason is that the set can consist of the whole parameter space and yet the confidence is γ = 1 – α in any case. In this paper, we study the problem of constructing data-dependent estimators better than γ₊, A reasonable estimator appears to be γ₊, which is one if C_F is the whole parameter space and γ otherwise. By using an estimated confidence approach and a squared-error loss, it is shown that γ₊ dominates γ. The risk improvement of γ₊ over γ can be sizable. Also, by numerically comparing γ₊ with a generalized Bayes estimator γ_L, which is shown to be admissible when one or two ratios are concerned, it is shown that γ₊ is nearly admissible. We also conclude that the common practice of reporting 1 – α only when C_F is not the whole parameter space is nearly admissible.     

State‐space model for proxy‐based millennial reconstruction

It is important to study historical temperature time series prior to the industrial revolution so that one can view the current global warming trend from a long‐term historical perspective. Because there are no instrumental records of such historical temperature data, climatologists have been interested in reconstructing historical temperatures using various proxy time series. In this paper, the authors examine a state‐space model approach for historical temperature reconstruction which not only makes use of the proxy data but also information on external forcings. A challenge in the implementation of this approach is the estimation of the parameters in the state‐space model. The authors developed two maximum likelihood methods for parameter estimation and studied the efficiency and asymptotic properties of the associated estimators through a combination of theoretical and numerical investigations. The Canadian Journal of Statistics 38: 488–505; 2010 © 2010 Crown in the right of Canada     

Mixture Distributions Based Methods of Calibration for the Empirical Log-Likelihood Ratio

Empirical likelihood ratio confidence regions based on the chi-square calibration suffer from an undercoverage problem in that their actual coverage levels tend to be lower than the nominal levels. The finite sample distribution of the empirical log-likelihood ratio is recognized to have a mixture structure with a continuous component on [0, + ∞) and a point mass at + ∞. The undercoverage problem of the Chi-square calibration is partly due to its use of the continuous Chi-square distribution to approximate the mixture distribution of the empirical log-likelihood ratio. In this article, we propose two new methods of calibration which will take advantage of the mixture structure; we construct two new mixture distributions by using the F and chi-square distributions and use these to approximate the mixture distributions of the empirical log-likelihood ratio. The new methods of calibration are asymptotically equivalent to the chi-square calibration. But the new methods, in particular the F mixture based method, can be substantially more accurate than the chi-square calibration for small and moderately large sample sizes. The new methods are also as easy to use as the chi-square calibration.     

NEIGHBORHOOD DISADVANTAGE AMONG RACIAL AND ETHNIC GROUPS:

We compare the neighborhood characteristics of native- and foreign-born blacks, whites, Hispanics, and Asians in 1970 and 1980. We broaden the locational attainment literature by emphasizing three contrasts: between black and nonblack groups, between native black and nonblack immigrant groups, and among black groups. Consistent with previous evidence, we find a clear spatial disadvantage for black groups relative to nonblack groups, and for native blacks compared to nonblack immigrant groups, in both years. However, our study reveals a slight advantage for foreign nonHispanic blacks (e.g., Afro-Caribbean immigrants) among the black groups throughout the time period. Our results break new ground by extending the analysis of racial and ethnic variation in residential attainment back to 1970, providing an earlier benchmark against which current patterns of residential attainment can be compared.