Robust model selection using the out-of-bag bootstrap in linear regression.

Outlying observations have a large influence on the linear model selection process. In this article, we present a novel approach to robust model selection in linear regression to accommodate the situations where outliers are present in the data. The model selection criterion is based on two components, the robust conditional expected prediction loss, and a robust goodness-of-fit with a penalty term. We estimate the conditional expected prediction loss by using the out-of-bag stratified bootstrap approach. In the presence of outliers, the stratified bootstrap ensures that we obtain bootstrap samples that are similar to the original sample data. Furthermore, to control the undue effect of outliers, we use the robust MM-estimator and a bounded loss function in the proposed criterion. Specifically, we observe that instead of minimizing the penalized loss function or the conditional expected prediction loss separately, it is better to minimize them simultaneously. The simulation and real-data based studies confirm the consistent and satisfactory behavior of our bootstrap model selection procedure in the presence of response outliers and covariate outliers.

Total retail goods consumption, industry structure, urban population growth and pollution intensity: an application of panel data analysis for China.

There has been a growing concern regarding the regulation of environmental pollution in the face of a growing population, global warming, and climate change. Governments around the world have devised various mechanisms and policy strategies to ameliorate the worsening condition of natural environment around the world. Similar to the developed world, in China, the government is also aware of deteriorating environmental conditions. Hence, the existing abatement instruments include pollution discharge fees and several other policy strategies. This research is conducted to investigate the association between pollution intensity and its determinants, i.e., pollutant discharge fees and urban population, third industry structure, and total retail goods consumption. The secondary data of 29 provinces is used for empirical analysis. The principal component analysis is used to develop a single index called pollution intensity, and panel autoregressive distributed lags model (ARDL), or pooled mean group (PMG) analysis, is employed to find long-run and short-run relationship. The empirical findings show that pollution discharge fees negatively affects pollution intensity. Total retail good consumption and urban population increase pollution intensity. However, third industry structure helps to control pollution intensity. These results suggest reforms in the existing environmental regulations policy by targeting more pollutant intensive provinces.

Spectrum of beta-thalassemia mutations in various regions of Punjab and Islamabad, Pakistan: establishment of prenatal diagnosis.

We present here an analysis of 888 unrelated beta-thal chromosomes consisting of 444 transfusion dependent children from various regions of Punjab and Islamabad Pakistan. By using Multiplex ARMS- PCR, restriction endonuclease analysis, allele specific oligonucleotide (ASO) hybridization and sequencing, 17 beta-thal mutations and 3 Hb variants were detected in 99.5 % (884/888) of the chromosomes analyzed. First trimester prenatal diagnosis by chorionic villus sampling (CVS) was also carried out in seven pregnancies at risk of beta-thalassemia. Our results indicate that three most common mutations accounted for 86.8% of the beta-thal alleles in this region. These findings have important implications for prevention of beta-thalassemia through genetic counseling and prenatal diagnosis in this part of Pakistan.