How can we report returns for a swap that has zero value? How can we perform return optimization for a zero-value long-short portfolio? By introducing a suitable "basis", it is possible to extend the definition of returns to leveraged products in such a way that performance attribution and...

If the distribution of a financial variable is highly non-normal, as is the case for the monthly return of some hedge funds or options, how do we compute the projected annualized skewness and kurtosis? We address this question in greater generality, projecting all the summary statistics of the...

The intuitive meaning of "beta" is well known to all risk and portfolio managers: the beta is the sensitivity of the return on a given asset to a given risk factor. The applications of the "beta" are manifold, from risk computation and analysis to hedging. However, the precise definition and...

We show how to mix machine learning signals such as kernel smoothing and fuzzy memberships via the Entropy Pooling approach by Meucci (2008). We illustrate a case study, where we overlay an exponentially time-decayed prior to a pseudo-Gaussian kernel that emphasizes market scenarios where...

Propagating causal stress-tests or contagion on selected risk factors to all the risk drivers is a challenging task. We use Entropy Pooling by Meucci (2008) to address this issue. Our causal stress-tests comprise, but are not restricted to, stress-testing Bayesian networks. We detail the theory...

We extend the Fully Flexible Views generalization of the Black-Litterman approach to effectively handle extreme views on the tails of a distribution. First, we provide a recursive algorithm to process views on the conditional value at risk, which cannot be handled directly by the original...

We introduce "factors on demand", a modular, multi-asset-class return decomposition framework that extends beyond the standard systematic-plus-idiosyncratic approach. This framework, which rests on the conditional link between flexible bottom-up estimation factor models and flexible top-down...

We present a simple method to generate scenarios from multivariate elliptical distributions where the sample mean and covariances match the respective population moments. This methodology easily applies to large numbers of scenarios and large-dimensional distributions. We show an application to...

We introduce the multivariate Ornstein-Uhlenbeck and discuss how it generalizes a vast class of continuous-time and discrete-time multivariate processes. Relying on the simple geometrical interpretation of the dynamics of the Ornstein-Uhlenbeck process we introduce cointegration and its...