probabilistic_sharpe_ratio

Defined in fynance.research

probabilistic_sharpe_ratio(sr, n_obs, *, sr_benchmark=0.0, skew=0.0, kurt=3.0)

Probabilistic Sharpe ratio (PSR).

The probability that the true Sharpe exceeds sr_benchmark given the estimate sr from n_obs observations, correcting for the returns’ skewness and kurtosis (Bailey & López de Prado).

Parameters:

srfloat

Observed (non-annualized) Sharpe ratio.

n_obsint

Number of return observations.

sr_benchmarkfloat

Benchmark Sharpe to beat.

skewfloat

Skewness of the returns.

kurtfloat

Kurtosis of the returns (3 for a normal distribution).

Returns:

float

PSR in [0, 1].