Empirical likelihood based longitudinal studies

In longitudinal data analysis, our primary interest is in the regression parameters for the marginal expectations of the longitudinal responses; the longitudinal correlation parameters are of secondary interest. The joint likelihood function for longitudinal data is challenging, particularly for correlated discrete outcome data. Marginal modeling approaches such as generalized estimating equations (GEEs) have received much attention in the context of longitudinal regression. These methods are based on the estimates of the first two moments of the data and the working correlation structure. The confidence regions and hypothesis tests are based on the asymptotic normality. The methods are sensitive to misspecification of the variance function and the working correlation structure. Because of such misspecifications, the estimates can be inefficient and inconsistent, and inference may give incorrect results. To overcome this problem, we propose an empirical likelihood (EL) procedure based on a set of estimating equations for the parameter of interest and discuss its characteristics and asymptotic properties. We also provide an algorithm based on EL principles for the estimation of the regression parameters and the construction of a confidence region for the parameter of interest. We extend our approach to variable selection for highdimensional longitudinal data with many covariates. In this situation it is necessary to identify a submodel that adequately represents the data. Including redundant variables may impact the model’s accuracy and efficiency for inference. We propose a penalized empirical likelihood (PEL) variable selection based on GEEs; the variable selection and the estimation of the coefficients are carried out simultaneously. We discuss its characteristics and asymptotic properties, and present an algorithm for optimizing PEL. Simulation studies show that when the model assumptions are correct, our method performs as well as existing methods, and when the model is misspecified, it has clear advantages. We have applied the method to two case examples.

Modelling the growth of Northern Cod

We develop a body size growth model of Northern cod (Gadus morhua) in Northwest Atlantic Fisheries Organization (NAFO) Divisions 2J3KL during 2009-2013. We use individual length-at-age data from the bottom trawl survey in these divisions during 2009–2013. We use the Von Bertalanffy (VonB) model extended to account for between-individual variations in growth, and variations that may be caused by the methods which fish are caught and sampled for length and age measurements. We assume between-individual variation in growth appears because individuals grow at a different rate (k), and they achieve different maximum sizes (l∞). We also included measurement error in length and age in our model since ignoring these errors can lead to biased estimates of the growth parameters. We use the structural errors-invariables (SEV) approach to estimate individual variation in growth, ageing error variation, and the true age distribution of the fish. Our results shows the existence of individual variation in growth and ME in age. According to the negative log likelihood ratio (NLLR) test, the best model indicated: 1) different growth patterns across divisions and years. 2) Between individual variation in growth is the same for the same division across years. 3) The ME in age and true age distribution are different for each year and division.