pfhedge.nn.functional.expected_shortfall¶

pfhedge.nn.functional.expected_shortfall(input, p, dim=None)[source]¶

Returns the expected shortfall of the given input tensor.

Parameters

input (torch.Tensor) – The input tensor.
p (float) – The quantile level.
dim (int, optional) – The dimension to sort along.

Returns

torch.Tensor

Examples

>>> from pfhedge.nn.functional import expected_shortfall
>>>
>>> input = -torch.arange(10.0)
>>> input
tensor([-0., -1., -2., -3., -4., -5., -6., -7., -8., -9.])
>>> expected_shortfall(input, 0.3)
tensor(8.)