Maximum Likelihood Estimation for the Pooled Repeated Partly Interval-Censored Observations Logistic Regression Model

Often in longitudinal studies, some subjects complete their follow-up visits, but others miss their visits due to various reasons. For those who miss follow-up visits, some of them might learn that the event of interest has already happened when they come back. In this case, not only are their event times interval-censored, but also their time-dependent measurements are incomplete. This problem was motivated by a national longitudinal survey of youth data. Maximum likelihood estimation (MLE) method based on expectation-maximization (EM) algorithm is used for parameter estimation. Then missing information principle is applied to estimate the variance-covariance matrix of the MLEs. Simulation studies demonstrate that the proposed method works well in terms of bias, standard error, and power for samples of moderate size. The national longitudinal survey of youth 1997 (NLSY97) data is analyzed for illustration.

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.

Study on a gravity precursor mode of the M_S=7.0 Lijiang earthquake

To study the seismogeny process or the precursory behavior of the 1976 MS=7.0 Lijiang earthquake, we analyze the repeat gravity data with high precision from the Western Yunnan Earthquake Prediction Experiment Area (WYEPEA) and the related results of geology and geophysics survey in this paper. Considering the gross errors caused by observation data and model difference, we have firstly inverted the slip distribution of the main active faults with time based on the robust Bayesian least squares estimation and multi-fault dislocation model. The re-sults show that the slip changes of the faults with time from 1990 to 1997 obviously reflect the preparation process of the Lijiang earthquake. The images of main precursor mode have the characteristic of main shock-after shock type, which is agreement to the model of coupling movement between crust density and crust deformation (DD mode of coupling movement).