Probabilistic Methods

pyvallocation.probabilities.compute_effective_number_scenarios(probabilities: numpy.ndarray) → float[source]: Return the effective number of scenarios given a probability vector.

pyvallocation.probabilities.generate_exp_decay_probabilities(num_observations: int, half_life: int) → numpy.ndarray[source]: Return exponentially decaying probabilities with the given half_life.

pyvallocation.probabilities.generate_gaussian_kernel_probabilities(x: numpy.ndarray, v: numpy.ndarray | None = None, h: float | None = None, x_T: float | None = None) → numpy.ndarray[source]: Generate kernel-based probabilities for v centred on x_T.

pyvallocation.probabilities.generate_uniform_probabilities(num_observations: int) → numpy.ndarray[source]: Return equal probabilities for num_observations scenarios.

pyvallocation.probabilities.silverman_bandwidth(x: numpy.ndarray) → float[source]: Return Silverman’s rule-of-thumb bandwidth for x.