qis.regimes.compute_regime_betas

qis.regimes.compute_regime_betas(sampled_returns_with_regime_id, benchmark, af, min_periods=24, regime_column='regime')[source]

Per-regime OLS betas of every asset on the benchmark, with the intercepts discarded.

Each asset uses the periods where it and the benchmark both have a return; the regimes are those of the sampled frame, set on the whole benchmark sample.

Parameters:

  • sampled_returns_with_regime_id (DataFrame) – periodic returns with a regime column

  • benchmark (str) – name of the benchmark column

  • af (float) – annualisation factor of the periodic returns, for the idiosyncratic volatility

  • min_periods (int) – fewest common periods an asset needs

  • regime_column (str) – name of the regime column

Returns:

one row per asset other than the benchmark and, per regime id in bucket order, beta_<id> and n_<id>, then beta_total and idio_vol, the annualised standard deviation of the pooled piecewise residuals

Raises:

ValueError – if an asset has fewer than min_periods common periods

Return type:

DataFrame