Regression III – A Practical Example

We’ve already covered the basics and some of the theory or regression. So now, let’s do a practical example of linear regression in python.

Let’s mock up some practice data. We’ll create a simple predictor X, that just takes all the values between 0 and 20 in steps of 0.1, and a response Y, which a depends on X linearly via Y = 5X + 3 + (random noise). We generate this in python like so

import numpy as np

X = np.arange(0, 20, 0.1)
Y = 5 * X + 3 + np.random(200)/10

Now we can create a linear model and fit it with scikit-learn

from sklearn.linear_model import LinearRegression
model = LinearRegression().fit(X.reshape(-1,1),Y)

We need to reshape the X array, because the fit method expects a two dimensional array, the outer dimension for the samples and the inner dimension is the predictors.

We can see the coefficients of this model with


Which should give something like

(array([5.00041147]), 3.0461878369249646)

Then, if we would like to forecast values, say for the X value 11.34, we can use the predict method on our model


Leave a Reply

Your email address will not be published. Required fields are marked *