I would like to calculate the inverse CDF of a Variable in pymc. I would typically use norm.ppf from Scipy, however within this function there are boolean operations and so I get the type error “Variables do not support boolean operations.” I would normally get around this error by using aesara.ife…

Thanks a lot! The the cdf function works, but I think the invcdf function needs a little shift, but this works for me now. def cdf(x): return 0.5 + 0.5 * pm.math.erf(x / pm.math.sqrt(2)) def invcdf(x): x = 0.5-x return -pm.math.sqrt(2) * pm.math.erfinv(2*x)

Percent point function (inverse of `CDF`) in pymc

Questions v5

Simon September 6, 2022, 10:07am 6

I see, thanks for the clarification

Topic		Replies	Views
Scipy.stats distributions cdfs Questions	1	1865	August 13, 2018
Custom likelihood with inverse CDF Questions	4	965	May 17, 2019
TypeError: Variables do not support boolean operations Questions	0	1711	March 17, 2020
Computiong the PPF (inverse CDF) for PYMC distributions v5 modeling	3	628	October 20, 2022
Boolean operations support for theano.tensor Questions theano	1	2427	July 13, 2017

Percent point function (inverse of `CDF`) in pymc

Related topics