Multivariate generalized Birnbaum—Saunders kernel density estimators

In this article, we first propose the classical multivariate generalized Birnbaum–Saunders kernel estimator for probability density function estimation in the context of multivariate non negative data. Then, we apply two multiplicative bias correction (MBC) techniques for multivariate kernel density estimator. Some properties (bias, variance, and mean integrated squared error) of the corresponding estimators are also investigated. Finally, the performances of the classical and MBC estimators based on family of generalized Birnbaum–Saunders kernels are illustrated by a simulation study.     

How might sovereign bond yields in Asia Pacific react to US monetary normalisation under turbulent market conditions?*

This paper examines the potential impact of US monetary normalisation on sovereign bond yields in Asia Pacific. We apply the quantile vector autoregressive model with principal component analysis to the assessment of tail risk of sovereign debt, which may not be detectable using traditional OLS-based analysis. Our empirical evidence suggests that US Treasury bond yields can have a significant impact on sovereign bond yields in the region, an important channel through which monetary normalisation by the Fed can affect Asia-Pacific economies. Increases in sovereign bond yields will not only compromise the ability of the sovereigns in the region to service their debt but also translate into higher costs of borrowing for the rest of economy. The results show how much the outsized impact could potentially be if US monetary normalisation somehow turns out to be much more disorderly than expected.     

Economic statistical design of X‾-control charts: Modified version of Rahim and Banerjee (1993) model

Rahim and Banerjee considered a constant integral of the hazard function for all sampling intervals. This led the sampling intervals to depend on the extended first sampling interval (h₁). Since this limitation might not lead to an optimal situation, we first showed that elimination of the mentioned restriction did not cause any significant change in the average quality cycle cost. So if one is looking for an ideal cost and the simplicity of the process, the approach taken in Rahim and Banerjee’s study is the best procedure to adopt. Moreover, in many cases of non-uniform sampling method the first sampling interval becomes so large and this can sometimes lead the production system to the out-of-control state due to unexpected failures that might happen during that time. Therefore, we proposed a new model of uniform and non-uniform sampling intervals combination that allows us to confine the value of h₁ without undergoing high costs. The proposed model showed that the quality cycle cost of the proposed model is lower than Rahim and Banerjee’s model in the economic-statistical state. For more illustration, we conducted sensitivity analysis and gave numerical examples.     

Approximate asymptotic Bahadur efficiency of independence tests with random sample size

We compare various kind of independence tests based on samples with random size, in order to provide practitioners with some guidance for their choice based on approximate Bahadur efficiency. Such results are obtained for a wide class of distributions of the random index; the efficiency slopes of the statistics we consider are then expressed in terms of the parameters of these same distributions.     

Quantile regression in heteroscedastic varying coefficient models

Varying coefficient models are flexible models to describe the dynamic structure in longitudinal data. Quantile regression, more than mean regression, gives partial information on the conditional distribution of the response given the covariates. In the literature, the focus has been so far mostly on homoscedastic quantile regression models, whereas there is an interest in looking into heteroscedastic modelling. This paper contributes to the area by modelling the heteroscedastic structure and estimating it from the data, together with estimating the quantile functions. The use of the proposed methods is illustrated on real-data applications. The finite-sample behaviour of the methods is investigated via a simulation study, which includes a comparison with an existing method.     

Optimal sample size allocation for multi-level stress testing with Weibull regression under Type-II censoring

We discuss the optimal allocation problem in a multi-level stress test with Type-II censoring and Weibull (extreme value) regression model. We derive the maximum-likelihood estimators and their asymptotic variance–covariance matrix through the Fisher information. Four optimality criteria are used to discuss the optimal allocation problem. Optimal allocation of units, both exactly for small sample sizes and asymptotically for large sample sizes, for two- and four-stress-level situations are determined numerically. Conclusions and discussions are provided based on the numerical studies.     

Testing for heteroscedasticity in a dynamic simultaneous equation model

This paper generalizes the Lagrange multiplier test for heteroscedasticity to a dynamic simultaneous equation model. A proof ofthe validity of the test is given. Small sample proper- ties of the Lagrange multiplier test and its 'studentized' version, under normal and non-normal errors, are investigated using a Monte Carlo experiment.