fynance.features.metrics.roll_drawdown¶
-
fynance.features.metrics.
roll_drawdown
(X, w=None, raw=False, axis=0, dtype=None)¶ Measures the rolling drawdown of each X’ series.
Function to compute measure of the decline from a historical peak in some variable [5] (typically the cumulative profit or total open equity of a financial trading strategy).
Parameters: - X : np.ndarray[dtype, ndim=1 or 2]
Time-series of prices, performances or index. Must be positive values.
- w : int, optional
Size of the lagged window of the rolling function, must be positive. If
w is None
orw=0
, thenw=X.shape[axis]
. Default is None.- raw : bool, optional
- If True then compute the raw drawdown.
- Else (default) compute the drawdown in percentage.
- axis : {0, 1}, optional
Axis along wich the computation is done. Default is 0.
- dtype : np.dtype, optional
The type of the output array. If dtype is not given, infer the data type from X input.
Returns: - np.ndarray[dtype, ndim=1 or 2]
Series of drawdown for each series.
Notes
Let DD^w the drawdown vector with a lagged window of size w:
\[\begin{split}DD^w_t =\begin{cases} max(X_{t - w + 1:t}) - X_t \text{, if raw=True} \\ 1 - \frac{X_t}{max(X_{t - w + 1:t})} \text{, otherwise} \\ \end{cases}\end{split}\]References
[5] https://en.wikipedia.org/wiki/Drawdown_(economics) Examples
>>> X = np.array([70, 100, 80, 120, 160, 80]).astype(np.float64) >>> roll_drawdown(X) array([0. , 0. , 0.2, 0. , 0. , 0.5]) >>> roll_drawdown(X.reshape([6, 1])).T array([[0. , 0. , 0.2, 0. , 0. , 0.5]]) >>> roll_drawdown(X, raw=True) array([ 0., 0., 20., 0., 0., 80.]) >>> X = np.array([100, 80, 70, 75, 110, 80]).astype(np.float64) >>> roll_drawdown(X, raw=True, w=3) array([ 0., 20., 30., 5., 0., 30.])