Student

Models for regression coefficients. See Chapter 12 in Regression and Other Stories.

Source: Student/student.Rmd.

The Python version keeps data handling explicit and uses lapylace for Stan-backed generalized linear models, so the statistical model can be read from a formula rather than from handwritten Stan.

from pathlib import Path
import sys
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import lapylace as lp

sys.path.append(str(Path.cwd().parent / 'python'))
from data import ros_path

Data

student_merged = pd.read_csv(ros_path('Student/data', 'student-merged.csv'))
student_merged.head()

	G1mat	G2mat	G3mat	G1por	G2por	G3por	age	...	higher	internet	romantic	famrel	freetime	goout	Dalc	Walc	health	absences
0	7	10	10	13	13	13	15	...	1	1	0	3	1	2	1	1	1	2
1	8	6	5	13	11	11	15	...	1	1	1	3	3	4	2	4	5	2
2	14	13	13	14	13	12	15	...	1	0	0	4	3	1	1	1	2	8
3	10	9	8	10	11	10	15	...	1	1	0	4	3	2	1	1	5	2
4	10	10	10	13	13	13	15	...	1	1	1	4	2	1	2	3	3	8

5 rows × 32 columns

Models

The formulas below are the Python counterparts of the model formulas in the source example. Use lapylace for the Stan-backed Bayesian fit with the same formula interface.

# fit = lp.stan_glm('G3mat ~ _', data=student_merged, family=lp.gaussian(), chains=4, iter_sampling=1000)

# fit = lp.stan_glm('G3mat ~ _', data=student_merged, family=lp.gaussian(), chains=4, iter_sampling=1000)

# fit = lp.stan_glm('G3mat ~ _', data=student_merged, family=lp.gaussian(), chains=4, iter_sampling=1000)

# fit = lp.stan_glm('G3mat ~ _', data=student_merged, family=lp.gaussian(), chains=4, iter_sampling=1000)

# fit = lp.stan_glm('G3mat ~ failures + schoolsup + goout + absences', data=student_merged, family=lp.gaussian(), chains=4, iter_sampling=1000)

Notes

Source computation blocks represented: 41.
Data paths are expressed through the shared ros_path() helper.
Formula-based Bayesian regressions are routed through lapylace.stan_glm().
Plotting and simulation work uses NumPy, pandas, matplotlib idioms.