|
|
Matthias Knab, Opalesque for New Managers: A new research note by Francois-Serge Lhabitant offers a sharp critique of how the asset management industry estimates and interprets alpha, arguing that many widely used methodologies systematically distort the true picture of manager skill. The paper shows that alpha - typically the intercept from a regression of fund excess returns on a benchmark - is far more fragile, model-dependent and statistically uncertain than most practitioners appreciate.
Across both academic models and real hedge-fund examples, the study finds that alpha is often an artefact of omitted risks, imperfect benchmarks, unstable betas and inappropriate assumptions about timing behaviour. In practice, the author argues, the "alpha" many investors believe they are paying for is frequently a statistical mirage.
The core problem: alpha depends on the model, not the manager
The paper shows that even small changes in the way alpha is calculated - such as using total returns instead of excess returns, or selecting one benchmark instead of another - can dramatically alter the estimated outperformance. Standard methods also incorrectly treat alpha as a fixed, unconditional number, even though true risk exposures vary over market regimes.
Several structural issues distort alpha estimates:
|
|
RSS
