sergiosonline

#!/usr/bin/env python
# coding: utf-8

# # Intro to PyStan

# Stan is a computation engine for Bayesian model fitting. It relies on HMC to sample from the posterior distribution of the desired model.
# 
# Here are the detailed installation steps to set up Stan: https://pystan.readthedocs.io/en/latest/installation_beginner.html
# 
# For MacOS:

jhconning

jhconning

rudeboybert

# R source code for all slides/videos in Albert Y. Kim's "Modeling with Data in
# the Tidyverse" DataCamp course:
# https://www.datacamp.com/courses/modeling-with-data-in-the-tidyverse
# This code is available at http://bit.ly/modeling_tidyverse

# Load all necessary packages -----
library(ggplot2)
library(dplyr)
library(moderndive)

kgori

import scipy
import scipy.stats as ss
import numpy as np

def discretize(alpha, ncat, dist=ss.gamma):
    if dist == ss.gamma:
        dist = dist(alpha, scale=1 / alpha)
    elif dist == ss.lognorm:
        dist = dist(s=alpha, scale=np.exp(0.5 * alpha**2))
    quantiles = dist.ppf(np.arange(0, ncat) / ncat)