In this tutorial, you’ll see how to perform multiple linear regression in Python using both sklearn and statsmodels.
Here are the topics to be covered:
- Reviewing the example to be used in this tutorial
- Checking for Linearity
- Performing the multiple linear regression in Python
Example of Multiple Linear Regression in Python
In the following example, we will perform multiple linear regression for a fictitious economy, where the index_price is the dependent variable, and the 2 independent/input variables are:
- interest_rate
- unemployment_rate
Please note that you will have to validate that several assumptions are met before you apply linear regression models. Most notably, you have to make sure that a linear relationship exists between the dependent variable and the independent variable/s (more on that under the checking for linearity section).
Let’s now jump into the dataset that we’ll be using. The data will be captured using Pandas DataFrame:
import pandas as pd data = {'year': [2017,2017,2017,2017,2017,2017,2017,2017,2017,2017,2017,2017,2016,2016,2016,2016,2016,2016,2016,2016,2016,2016,2016,2016], 'month': [12,11,10,9,8,7,6,5,4,3,2,1,12,11,10,9,8,7,6,5,4,3,2,1], 'interest_rate': [2.75,2.5,2.5,2.5,2.5,2.5,2.5,2.25,2.25,2.25,2,2,2,1.75,1.75,1.75,1.75,1.75,1.75,1.75,1.75,1.75,1.75,1.75], 'unemployment_rate': [5.3,5.3,5.3,5.3,5.4,5.6,5.5,5.5,5.5,5.6,5.7,5.9,6,5.9,5.8,6.1,6.2,6.1,6.1,6.1,5.9,6.2,6.2,6.1], 'index_price': [1464,1394,1357,1293,1256,1254,1234,1195,1159,1167,1130,1075,1047,965,943,958,971,949,884,866,876,822,704,719] } df = pd.DataFrame(data) print(df)
Here is the full dataset:
year month interest_rate unemployment_rate index_price
0 2017 12 2.75 5.3 1464
1 2017 11 2.50 5.3 1394
2 2017 10 2.50 5.3 1357
3 2017 9 2.50 5.3 1293
4 2017 8 2.50 5.4 1256
5 2017 7 2.50 5.6 1254
6 2017 6 2.50 5.5 1234
7 2017 5 2.25 5.5 1195
8 2017 4 2.25 5.5 1159
9 2017 3 2.25 5.6 1167
10 2017 2 2.00 5.7 1130
11 2017 1 2.00 5.9 1075
12 2016 12 2.00 6.0 1047
13 2016 11 1.75 5.9 965
14 2016 10 1.75 5.8 943
15 2016 9 1.75 6.1 958
16 2016 8 1.75 6.2 971
17 2016 7 1.75 6.1 949
18 2016 6 1.75 6.1 884
19 2016 5 1.75 6.1 866
20 2016 4 1.75 5.9 876
21 2016 3 1.75 6.2 822
22 2016 2 1.75 6.2 704
23 2016 1 1.75 6.1 719
Checking for Linearity
Before you execute a linear regression model, it is advisable to validate that certain assumptions are met.
As noted earlier, you may want to check that a linear relationship exists between the dependent variable and the independent variable/s.
In our example, you may want to check that a linear relationship exists between the:
- index_price (dependent variable) and interest_rate (independent variable)
- index_price (dependent variable) and unemployment_rate (independent variable)
To perform a quick linearity check, you can use scatter diagrams (utilizing the matplotlib library). For example, you can use the code below in order to plot the relationship between the index_price and the interest_rate:
import pandas as pd import matplotlib.pyplot as plt data = {'year': [2017,2017,2017,2017,2017,2017,2017,2017,2017,2017,2017,2017,2016,2016,2016,2016,2016,2016,2016,2016,2016,2016,2016,2016], 'month': [12,11,10,9,8,7,6,5,4,3,2,1,12,11,10,9,8,7,6,5,4,3,2,1], 'interest_rate': [2.75,2.5,2.5,2.5,2.5,2.5,2.5,2.25,2.25,2.25,2,2,2,1.75,1.75,1.75,1.75,1.75,1.75,1.75,1.75,1.75,1.75,1.75], 'unemployment_rate': [5.3,5.3,5.3,5.3,5.4,5.6,5.5,5.5,5.5,5.6,5.7,5.9,6,5.9,5.8,6.1,6.2,6.1,6.1,6.1,5.9,6.2,6.2,6.1], 'index_price': [1464,1394,1357,1293,1256,1254,1234,1195,1159,1167,1130,1075,1047,965,943,958,971,949,884,866,876,822,704,719] } df = pd.DataFrame(data) plt.scatter(df['interest_rate'], df['index_price'], color='red') plt.title('Index Price Vs Interest Rate', fontsize=14) plt.xlabel('Interest Rate', fontsize=14) plt.ylabel('Index Price', fontsize=14) plt.grid(True) plt.show()
You’ll notice that indeed a linear relationship exists between the index_price and the interest_rate. Specifically, when interest rates go up, the index price also goes up.
And for the second case, you can use this code in order to plot the relationship between the index_price and the unemployment_rate:
import pandas as pd import matplotlib.pyplot as plt data = {'year': [2017,2017,2017,2017,2017,2017,2017,2017,2017,2017,2017,2017,2016,2016,2016,2016,2016,2016,2016,2016,2016,2016,2016,2016], 'month': [12,11,10,9,8,7,6,5,4,3,2,1,12,11,10,9,8,7,6,5,4,3,2,1], 'interest_rate': [2.75,2.5,2.5,2.5,2.5,2.5,2.5,2.25,2.25,2.25,2,2,2,1.75,1.75,1.75,1.75,1.75,1.75,1.75,1.75,1.75,1.75,1.75], 'unemployment_rate': [5.3,5.3,5.3,5.3,5.4,5.6,5.5,5.5,5.5,5.6,5.7,5.9,6,5.9,5.8,6.1,6.2,6.1,6.1,6.1,5.9,6.2,6.2,6.1], 'index_price': [1464,1394,1357,1293,1256,1254,1234,1195,1159,1167,1130,1075,1047,965,943,958,971,949,884,866,876,822,704,719] } df = pd.DataFrame(data) plt.scatter(df['unemployment_rate'], df['index_price'], color='green') plt.title('Index Price Vs Unemployment Rate', fontsize=14) plt.xlabel('Unemployment Rate', fontsize=14) plt.ylabel('Index Price', fontsize=14) plt.grid(True) plt.show()
You’ll notice that a linear relationship also exists between the index_price and the unemployment_rate – when the unemployment rates go up, the index price goes down (here we still have a linear relationship, but with a negative slope).
Next, we are going to perform the actual multiple linear regression in Python.
Performing the Multiple Linear Regression
Once you added the data into Python, you may use either sklearn or statsmodels to get the regression results.
Either method would work, but let’s review both methods for illustration purposes.
You may then copy the code below into Python:
import pandas as pd from sklearn import linear_model import statsmodels.api as sm data = {'year': [2017,2017,2017,2017,2017,2017,2017,2017,2017,2017,2017,2017,2016,2016,2016,2016,2016,2016,2016,2016,2016,2016,2016,2016], 'month': [12,11,10,9,8,7,6,5,4,3,2,1,12,11,10,9,8,7,6,5,4,3,2,1], 'interest_rate': [2.75,2.5,2.5,2.5,2.5,2.5,2.5,2.25,2.25,2.25,2,2,2,1.75,1.75,1.75,1.75,1.75,1.75,1.75,1.75,1.75,1.75,1.75], 'unemployment_rate': [5.3,5.3,5.3,5.3,5.4,5.6,5.5,5.5,5.5,5.6,5.7,5.9,6,5.9,5.8,6.1,6.2,6.1,6.1,6.1,5.9,6.2,6.2,6.1], 'index_price': [1464,1394,1357,1293,1256,1254,1234,1195,1159,1167,1130,1075,1047,965,943,958,971,949,884,866,876,822,704,719] } df = pd.DataFrame(data) x = df[['interest_rate','unemployment_rate']] y = df['index_price'] # Using sklearn regr = linear_model.LinearRegression() regr.fit(x, y) print('Intercept: \n', regr.intercept_) print('Coefficients: \n', regr.coef_) # Using statsmodels x = sm.add_constant(x) # adding a constant model = sm.OLS(y, x).fit() predictions = model.predict(x) print_model = model.summary() print(print_model)
Once you run the code in Python, you’ll observe two parts:
(1) The first part shows the output generated by sklearn:
Intercept:
1798.4039776258564
Coefficients:
[ 345.54008701 -250.14657137]
This output includes the intercept and coefficients. You can use this information to build the multiple linear regression equation as follows:
index_price = (intercept) + (interest_rate coef)*X1 + (unemployment_rate coef)*X2
And once you plug the numbers:
index_price = (1798.4040) + (345.5401)*X1 + (-250.1466)*X2
(2) The second part displays a comprehensive table with statistical info generated by statsmodels.
This information can provide you additional insights about the model used (such as the fit of the model, standard errors, etc):
OLS Regression Results
==============================================================================
Dep. Variable: index_price R-squared: 0.898
Model: OLS Adj. R-squared: 0.888
Method: Least Squares F-statistic: 92.07
Date: Sat, 30 Jul 2022 Prob (F-statistic): 4.04e-11
Time: 13:47:01 Log-Likelihood: -134.61
No. Observations: 24 AIC: 275.2
Df Residuals: 21 BIC: 278.8
Df Model: 2
Covariance Type: nonrobust
=====================================================================================
coef std err t P>|t| [0.025 0.975]
-------------------------------------------------------------------------------------
const 1798.4040 899.248 2.000 0.059 -71.685 3668.493
interest_rate 345.5401 111.367 3.103 0.005 113.940 577.140
unemployment_rate -250.1466 117.950 -2.121 0.046 -495.437 -4.856
==============================================================================
Omnibus: 2.691 Durbin-Watson: 0.530
Prob(Omnibus): 0.260 Jarque-Bera (JB): 1.551
Skew: -0.612 Prob(JB): 0.461
Kurtosis: 3.226 Cond. No. 394.
==============================================================================
Notice that the coefficients captured in this table (highlighted in yellow) match with the coefficients generated by sklearn.
That’s a good sign! we got consistent results by applying both sklearn and statsmodels.
Conclusion
Linear regression is often used in Machine Learning. You have seen some examples of how to perform multiple linear regression in Python using both sklearn and statsmodels.
Before applying linear regression models, make sure to check that a linear relationship exists between the dependent variable (i.e., what you are trying to predict) and the independent variable/s (i.e., the input variable/s).