MAXIMUM SCORE CHANGE-POINT ESTIMATION IN BINARY RESPONSE MODEL

This paper is concerned with the estimation of change-point in a binary response model with the assumption that the conditional median of the error term, given the explanatory variable, is zero. We construct an estimation of change-point based on the maximum score function and give its exponential convergence rate under some mild conditions.

Application of GEE Models for Assessing Maternal Health Complications

Bangladesh, a developing country, gained success towards the fifth-millennium development goals target of reducing its maternal mortality ratio by three quarters by 2015, but yet worked more on it for further reduction of maternal mortality. In this light, though Bangladesh is committed to the sustainable development goals target of reducing its maternal mortality ratio to be reduced from 170 to 105 per 100,000 live births, the scope of research on this issue is limited because the maternal morbidity data is scarce in Bangladesh. In this paper, the prospective data on maternal morbidity in rural Bangladesh (collected by BIRPERHT) have been employed to trace out the high-risk and life-threatening factors associated with pregnancy-related complications. The subject-specific generalized estimating equations (SS-GEE) model with random effect structure is used for multivariate binary data for the repeated observations. The findings indicate that the risk of suffering from pregnancy complications is higher for high economic status, lower age at marriage, not visited for medical check-ups, outside home workers, and having miscarriage or abortion. Comparing the SS-GEE model with other correlation structures and relative efficiency factors, the SS-GEE model with random effect structure is well fitted for the prospective repeated observation data.

CONFIDENCE LOWER LIMITS FOR RESPONSE PROBABILITIES UNDER THE LOGISTIC RESPONSE MODEL

The lower confidence limits for response probabilities based on binary response data under the logistic response model are considered by saddlepoint approach.The high order approximation to the conditional distribution of a statistic for an interested parameter and then the lower confidence limits of response probabilities are derived.A simulation comparing these lower confidence limits with those obtained from the asymptotic normality is conducted.The proposed approximation is applied to two real data sets.Numerical results show that the saddlepoint approximations are much more accurate than the asymptotic normality approximations,especially for the cases of small or moderate sample sizes.

Efficient Robbins–Monro procedure for multivariate binary data

This paper considers the problem of jointly estimating marginal quantiles of a multivariatedistribution.A sufficient condition for an estimator that converges in probability under a multivariate version of Robbins–Monro procedure is provided.We propose an efficient procedurewhich incorporates the correlation structure of the multivariate distribution to improve the estimation especially for cases involving extreme marginal quantiles.Estimation efficiency of theproposed method is demonstrated by simulation in comparison with a general multivariate Robbins–Monro procedure and an efficient Robbins–Monro procedure that estimates the marginalquantiles separately.

Regression models of Pearson correlation coefficient

We propose two simple regression models of Pearson correlation coefficient of two normal responses or binary responses to assess the effect of covariates of interest.Likelihood-based inference is established to estimate the regression coefficients,upon which bootstrap-based method is used to test the significance of covariates of interest.Simulation studies show the effectiveness of the method in terms of type-I error control,power performance in moderate sample size and robustness with respect to model mis-specification.We illustrate the application of the proposed method to some real data concerning health measurements.

An Adaptive Design to Assess the Reliability of the Pyrotechnic Control Subsystem in Opening the Solar Array

The pyrotechnic control subsystem plays an important role in opening the solar array of a satellite. Assessing the reliability of the subsystem requires determining the level of a control factor that is needed to cause the desired response and energy output with high probability. A two-phase adaptive design to estimate the level of interest is proposed and studied. The convergence of the design is obtained. A simulation study shows that the estimate is very close to its population value and is robust to the initial guess of the design. As an application, the design is used to assess the reliability of a real pyrotechnic control subsystem.