This article introduces a very flexible framework for causal and predictive market views and stress-testing. The framework elegantly combines Bayesian networks (BNs) and Entropy Pooling (EP). In the new framework, BNs are used to generate a finite set of joint causal views / stress-tests for the...

We build an equilibrium model to explain why stock return predictability concentrates in bad times. The key feature is that investors use different forecasting models, and hence assess uncertainty differently. As economic conditions deteriorate, uncertainty rises and investors' opinions...

We consider a canonical asset pricing model, where agents with quadratic preferences are allowed to retrade a limited set of securities over multiple periods, after which these securities expire, and agents consume their liquidation values. A key assumption in this model is that agents have...

The approximate agents' wealth and price invariant densities of the prediction market model presented in Kets et al.(2014) is derived using the Fokker-Planck equation of the associated continuous-time jump process. We show that the approximation obtained from the evolution of log-wealth...

We investigate market selection and bet pricing in a simple Arrow security economy which we show is equivalent to the repeated prediction market models studied in the literature. We derive the condition for long run survival of more than one agent (the crowd) and quantify the information content...

We provide empirical evidence that CAPM-betas positively predict asset returns when market returns are predicted to be high, which occurs about every other month. Consequently, the product of beta and the predicted market return (CAPM) predicts asset returns by combining the out-of-sample...

We investigate the dynamic problem of how much attention an investor should pay to news in order to learn about stock-return predictability and maximize expected lifetime utility. We show that the optimal amount of attention is U-shaped in the return predictor, increasing with both uncertainty...

We investigate the use of machine learning techniques into building statistically stable systematic allocation strategies. Traditionally, allocation processes usually rely on variations of Markowitz framework such as Mean Variance allocation, Maximum Diversity, Risk Allocation , Value at Risk,...

Dynamic equilibrium models based on present value computation imply that returns are predictable but also generate particular patterns of predictability in asset returns. I take advantage of this to construct a set of tests of Equilibrium Generated Predictability (EGP). I apply the tests to...

Value and momentum returns and combinations of them are explained by their loadings on global macroeconomic risk factors across both countries and asset classes. These loadings describe why value and momentum have positive return premia while at the same time being negatively correlated. The...