Building Pandas OLS Linear Regression Results

How would I build linear regression results for this linear regression made by me from pandas?

import pandas as pd from pandas.stats.api import ols df = pd.read_csv('Samples.csv', index_col=0) control = ols(y=df['Control'], x=df['Day']) one = ols(y=df['Sample1'], x=df['Day']) two = ols(y=df['Sample2'], x=df['Day']) 

I tried plot() but it did not work. I want to build all three samples on one graph, is there a pandas code or matplotlib code for hadle data in the format of these summaries?

In any case, the results look like this:

Control

 ------------------------Summary of Regression Analysis------------------------- Formula: Y ~ <x> + <intercept> Number of Observations: 7 Number of Degrees of Freedom: 2 R-squared: 0.5642 Adj R-squared: 0.4770 Rmse: 4.6893 F-stat (1, 5): 6.4719, p-value: 0.0516 Degrees of Freedom: model 1, resid 5 -----------------------Summary of Estimated Coefficients------------------------ Variable Coef Std Err t-stat p-value CI 2.5% CI 97.5% -------------------------------------------------------------------------------- x -0.4777 0.1878 -2.54 0.0516 -0.8457 -0.1097 intercept 41.4621 2.9518 14.05 0.0000 35.6766 47.2476 ---------------------------------End of Summary--------------------------------- 

one

 -------------------------Summary of Regression Analysis------------------------- Formula: Y ~ <x> + <intercept> Number of Observations: 6 Number of Degrees of Freedom: 2 R-squared: 0.8331 Adj R-squared: 0.7914 Rmse: 2.0540 F-stat (1, 4): 19.9712, p-value: 0.0111 Degrees of Freedom: model 1, resid 4 -----------------------Summary of Estimated Coefficients------------------------ Variable Coef Std Err t-stat p-value CI 2.5% CI 97.5% -------------------------------------------------------------------------------- x -0.4379 0.0980 -4.47 0.0111 -0.6300 -0.2459 intercept 29.6731 1.6640 17.83 0.0001 26.4116 32.9345 ---------------------------------End of Summary--------------------------------- 

two

 -------------------------Summary of Regression Analysis------------------------- Formula: Y ~ <x> + <intercept> Number of Observations: 5 Number of Degrees of Freedom: 2 R-squared: 0.8788 Adj R-squared: 0.8384 Rmse: 1.0774 F-stat (1, 3): 21.7542, p-value: 0.0186 Degrees of Freedom: model 1, resid 3 -----------------------Summary of Estimated Coefficients------------------------ Variable Coef Std Err t-stat p-value CI 2.5% CI 97.5% -------------------------------------------------------------------------------- x -0.2399 0.0514 -4.66 0.0186 -0.3407 -0.1391 intercept 24.0902 0.9009 26.74 0.0001 22.3246 25.8559 ---------------------------------End of Summary--------------------------------- 
+5
source share
1 answer

You may find this my question useful Getting a regression line to build a Pandas regression

I tried to find some of my code making an ols plot with Pandas, but couldn't lay my hands on it. In general, you will probably be better off using Statsmodels for this, it knows about Pandas data structures .. so the transition is not too complicated. Then my answer and links to examples will make more sense.

See also: http://nbviewer.ipython.org/gist/dartdog/9008026

+4
source

Source: https://habr.com/ru/post/1015015/


All Articles