Observed choices in Social Dilemma Games usually take the form of bounded integers. We propose a doubly-truncated count data framework to process such data. We compare this framework to past approaches based on ordered outcomes and truncated continuous densities using Bayesian estimation and model selection techniques. We find that all three frameworks (i) support the presence of unobserved het...

We investigate the role of initial conditions at colonial independence on economic growth in Africa in the post-independence period using Bayesian Model Averaging (BMA). A key innovation in our estimation methodology is that we incorporate parameter heterogeneity in model averaging as well as try to mitigate the endogeneity problem present in growth regressions. In order to ensure that differen...

Moody’s databases of corporate issuers of long term bonds and structured products are used to estimate asset correlations across a group of sectors, world regions and products. The estimation of a dynamic factor model for default risk is performed using Bayesian methods. Results indicate that a two factor model rather than the one factor model, as proposed by the Basel II framework, better repr...

A large body of evidence has emerged in recent studies confirming that macroeconomic factors play an important role in determining investor risk premia and the ultimate path of equity returns. This paper illustrates how widely tested financial and economic variables from these studies can be employed in a time varying dynamic sector allocation model for U.S. equities. The model developed here i...

This paper investigates the effect of a positive technology shock on per capita hours worked within the class of Bayesian Vector Auto-Regressive [BVAR] models. Such a framework avoids the current debate regarding the specification issue of per capita hours [level versus first-difference stationary]. Six priors are considered and for each, we examine the impulse responses of per capita hours fol...

We derive a PAC-Bayesian generalization bound for density estimation. Similar to the PAC-Bayesian generalization bound for classification, the result has the appealingly simple form of a tradeoff between empirical performance and the KL-divergence of the posterior from the prior. Moreover, the PACBayesian generalization bound for classification can be derived as a special case of the bound for ...

This paper addresses the problem of supervised classification using general Bayesian learning. General Bayesian learning consists of estimating the unknown class-conditional densities from a set of labelled samples. However, the estimation requires to evaluate intractable multidimensional integrals. This paper studies an implementation of general Bayesian learning based on MCMC methods.

We apply Pires’s coherence property between unconditional and conditional preferences that admit a CEU representation. In conjunction with consequentialism (only those outcomes on states which are still possible can matter for conditional preference) this implies that the conditional preference may be obtained from the unconditional preference by taking the Full Bayesian Update of the capacity....

We consider the problem of online classification in nonstationary environments. Specifically, we take a Bayesian approach to sequential parameter estimation of a logistic MCS, and compare this method with other algorithms for nonstationary classification. We comment on several design considerations.

It is well-know that estimation by reduced rank regression is given by the solution to a generalized eigenvalue problem. This paper presents a new proof to establish this result and provides additional insight into the structure of the estimation problem. The proof is a direct algebraic proof that some might find more intuitive than existing proofs. JEL Classification: C3, C32