mbapy.sats.reg

This function performs linear regression on a given DataFrame.

Function

linear_reg(x:str, y:str, df:pd.DataFrame) -> dict

Perform linear regression on the given DataFrame.

Parameters

• x (str): The column name for the independent variable.
• y (str): The column name for the dependent variable.
• df (pd.DataFrame): The DataFrame containing the data.

Returns

• dict: A dictionary containing the regression model, slope, intercept, and R-squared value.
  • 'regressor' (LinearRegression): The fitted regression model.
  • 'a' (float): The slope of the regression line.
  • 'b' (float): The intercept of the regression line.
  • 'r2' (float): The R-squared value of the regression.

Example

import pandas as pd
from sklearn.linear_model import LinearRegression

df = pd.DataFrame({'x': [1, 2, 3, 4, 5], 'y': [2, 4, 6, 8, 10]})
result = linear_reg('x', 'y', df)
print(result)

Output:

{
    'regressor': LinearRegression(),
    'a': 2.0,
    'b': 0.0,
    'r2': 1.0
}

Notes: - The function uses the LinearRegression class from the sklearn.linear_model module to perform the linear regression. - The independent variable x and dependent variable y should be column names in the DataFrame df. - The function reshapes the input arrays x and y to have a single feature. - The function returns a dictionary containing the fitted regression model, slope, intercept, and R-squared value.