Prior Specification in Bayesian Model Averaging: An application to Economic Growth
Journal Title: Annals. Computer Science Series - Year 2018, Vol 16, Issue 1
Abstract
Some recent cross-country cross-sectional analyses have employed Bayesian Model Averaging to tackle the issue of model uncertainty. Bayesian model averaging has become an important tool in empirical settings with large numbers of potential regressors and relatively limited number of observations. We examine the effect of a variety of prior assumptions on the inference, posterior inclusion probabilities of regressors and on predictive performance. Bayesian model averaging (BMA) has become a widely accepted way of accounting for model uncertainty in regression models. However, to implement BMA, a prior is usually specified in two parts: prior for the regression parameters and prior over the model space. Hence, the choice of prior specification becomes paramount in Bayesian inference, unfortunately, in practice, most Bayesian analyses are performed with the so-called non-informative priors (i.e. priors constructed by some formal rule). The arbitrariness in the choice of prior or choosing inappropriate priors often lead to badly behaved posteriors. It is therefore imperative to study the effect of choice of priors in Bayesian model averaging. Six candidate parameter priors namely, Unit information prior (UIP), Risk inflation criterion (RIC), Bayesian Risk Inflation criterion (BRIC), Hannan-Quinn criterion (HQ), Empirical Bayes (EBL) and hyper-g and three model priors: uniform, beta-binomial and binomial were examined in this study. The performances of the resulting eighteen cases were judged using posterior inference, posterior inclusion probabilities of regressors and predictive performance. Analyses were carried out using datasets with 8-potential drivers of growth for 126 countries from 2010 to 2014. Finally, our analysis shows that the EBL parameter prior with random model prior robustly identifies far more growth determinants than other priors.
Authors and Affiliations
Tayo P. OGUNDUNMADE, Adedayo A. ADEPOJU
Keywords
Related Articles
- EP ID EP540182
- DOI -
- Views 55
- Downloads 0
Development and Optimization of a Multimedia Product<br />
This article presents a new concept of a multimedia interactive product. It is a multi-user versatile platform that can be used for different purposes. The first implementation of the platform is a multi-player game call...
How does the Shift-insertion sort behave when the sorting elements follow a Normal distribution?
The present paper examines the behavior of Shift-insertion sort (insertion sort with shifting) for normal distribution inputs and is in continuation of our earlier work on this new algorithm for discrete distribution inp...
Handling Multicollinearity; A Comparative Study Of The Prediction Performance Of Some Methods Based On Some Probabiltiy Distributions
This study used some probability distribution (Gamma, Beta and Chi-square distributions) to assess the performance of partial least square regression (PLSR), ridge regression (RR) and LASSO regression (LR) methods. Ordin...
On the proficient use of GEV distribution: a case study of subtropical monsoon region in India
The paper deals with the probabilistic estimates of extreme maximum rainfall (Annual basis) in the Ranchi, Jharkhand (India). Extreme Value Distribution family models are tried to capture the uncertainty of data and fina...
Motivating students in learning mathematics with GeoGebra
In this study, I researched three cases of educating mathematics with computer for the gifted students in Korea. The findings show that students were motivated to study math by various reasons (making their beautiful wor...