ST
StateTrace
Visual Quant & Low-Latency Systems Lab
GitHub
Curriculum/sharpe-ratio

Sharpe Ratio

statistics·L1 · combinator·stub
Replacesthe belief that Sharpe is a universal performance number.

Sharpe is a noisy estimator whose variance depends on sample size, return autocorrelation (Lo 2002), and tail fatness. A 1-year Sharpe 2.0 from monthly data and a 1-year Sharpe 2.0 from daily data are not the same evidence — the daily version has 22× more samples and a correspondingly tighter confidence interval.

Prerequisites(root concept)
Bridges
  • autocorrelation-biasshared failure mode
    The Lo 2002 correction: when daily returns are autocorrelated, the naive annualised-Sharpe formula (×√252) over-states the truth; the autocorrelation-adjusted version recovers the unbiased estimate.
  • fat-tails-and-evtshared measurement
    Sharpe assumes the second moment is well-defined. For fat-tailed return series (the empirical norm in finance), higher moments dominate; Sortino, Calmar, and tail-adjusted metrics give a different ranking.
Status

This concept is a node in the curriculum DAG. The full lab — page blocks, done state, references — has not been authored yet. The relations above describe where it sits in the graph.

Author at: content/concepts/sharpe-ratio/card.ts