The Bayesian elastic net regression

A Bayesian elastic net approach is presented for variable selection and coefficient estimation in linear regression models. A simple Gibbs sampling algorithm was developed for posterior inference using a location-scale mixture representation of the Bayesian elastic net prior for the regression coefficients. The penalty parameters are chosen through an empirical method that maximizes the data marginal likelihood. Both simulated and real data examples show that the proposed method performs well in comparison to the other approaches.     

Improving operational performance through knowledge exchange with customers

Drawing upon a knowledge-based view of the supply chain, this study analyses the antecedents and consequences of tacit and explicit knowledge transfer, from the supplier's standpoint. The proposed conceptual model is tested using survey data from a sample of 218 Canadian manufacturers. Results show that the exchange of tacit and explicit knowledge in the context of a relationship between a supplier and a customer has a positive impact on the supplier's operational performance. As expected, tacit knowledge exchange appears to have a stronger impact on performance than explicit knowledge exchange. Results also confirm the importance of the level of integration and inter-firm trust for the exchange of both types of knowledge. Although many studies of knowledge transfer in the supply chain have been published to date, no empirical study has specifically examined the simultaneous effect of both types of knowledge.     

Bayesian quantile regression for ordinal longitudinal data

Since the pioneering work by Koenker and Bassett [27], quantile regression models and its applications have become increasingly popular and important for research in many areas. In this paper, a random effects ordinal quantile regression model is proposed for analysis of longitudinal data with ordinal outcome of interest. An efficient Gibbs sampling algorithm was derived for fitting the model to the data based on a location-scale mixture representation of the skewed double-exponential distribution. The proposed approach is illustrated using simulated data and a real data example. This is the first work to discuss quantile regression for analysis of longitudinal data with ordinal outcome.