Hi, I am kind of a newbie to pymc and previously worked with stan and Turing. My goal is to implement a model which depends on many stimulus properties with pymc where two distributions are combined to a new one, which further serves as pobability estimate of the number of responses. Depending on t…

[image] uferc: combined = pm.Normal('combined',bias_distr+stim_distr) This is not going to work as PyMC does not yet provide functionality to combine distribution. However, since you are combining Normal distribution it makes it somewhat easier: # bias_distr = pm.Normal('bias_distr', m…

Combine two distributions to new one & scale parameter modifying probability depending on category

Questions version agnostic

ricardoV94 August 12, 2022, 10:56am 8

Did you perhaps intend:
p = pm.math.exp(pm.logcdf(combined, x))?

Logcdf gives you a value in the range (-inf, 0], whereas p should be in the range of [0, 1]

1 Like

Topic		Replies	Views
Sum of distributions Questions	1	718	November 3, 2020
How to force estimated variables into another distribution variable Questions	6	667	January 9, 2020
Combining posterior distributions Questions	0	803	July 17, 2020
Combining the outputs from two models trained using different data Questions	3	786	January 27, 2019
How to model product of multiple random variables Questions	8	4252	June 28, 2019

Combine two distributions to new one & scale parameter modifying probability depending on category

Related topics