Reading coef value from OLS regression results

Question

I use pandas and statsmodels to do linear regression. However, i can't find any possible way to read the results. the results are displayed but i need to do some further calculations using coef values. is there any possible way to store coef values into a new variable?

import statsmodels.api as sm
import numpy
ones = numpy.ones(len(x[0]))
t = sm.add_constant(numpy.column_stack((x[0], ones)))
for m in x[1:]:
    t = sm.add_constant(numpy.column_stack((m, t)))
results = sm.OLS(y, t).fit()

score 5 · Accepted Answer · answered Jul 25 '17 at 21:48

5

According to the docs, an instance of RegressionResults is returned.

You can see all the available attributes there.

Maybe you are interested in:

params

The linear coefficients that minimize the least squares criterion. This is usually called Beta for the classical linear model.

Example:

model = sm.OLS(Y,X)
results = model.fit()
print(results.params)

answered Jul 25 '17 at 21:48

sascha

32,238
6
68
110

1

```It didn't help``` is almost useless. Maybe you want to explain what the problem is. – sascha Jul 25 '17 at 21:59
1

there was a problem with numpy as i forgot to reverse all (np's) to numpy. sorry. Your answer is correct and it solved the problem. – HussainBiedouh Jul 25 '17 at 22:11
1

results.params is the answer, here are further examples http://www.statsmodels.org/dev/examples/notebooks/generated/ols.html – HussainBiedouh Jul 25 '17 at 22:11

score -1 · Answer 2 · edited Oct 30 '22 at 15:22

-1

Try this:

B0, B1, B2, B3 = modelo.params

edited Oct 30 '22 at 15:22

ah bon

9,293
12
65
148

answered May 04 '22 at 19:03

lcarboneto

1

1

Code-only answers may resolve the problem, but do not help people with this problem in understanding why this resolves the issue. Could you please edit your answer and include details on why this code resolves the problem? – Marc Sances Oct 31 '22 at 09:09

Reading coef value from OLS regression results

2 Answers2