Question

Using Python Estimate the 5^th and the 2^nd degree polynomial regressions for this data:

Age  Salary
25  135000
55  260000
27  105000
35  220000
60  240000
65  265000
45  270000
40  300000
50  265000
30  105000

Please show the coding used.

Answer 1

#CODE

import numpy as np
import matplotlib.pyplot as plt
import pandas as pd

data = pd.read_csv('AgeSalary.csv')
print(data)

X = np.array(data['Age']).reshape(1,-1).T
Y = data['Salary'].values

## Without any polynomial transforming

from sklearn.linear_model import LinearRegression
reg = LinearRegression()
reg.fit(X,Y)

#plot the actual and predicted
plt.scatter(X,Y,color='red')
plt.plot(X,reg.predict(X),color='blue')
plt.xlabel('Age')
plt.ylabel("Salary")

## Using 5th degree polynomial

from sklearn.preprocessing import PolynomialFeatures
poly5 = PolynomialFeatures(degree = 5)
X_poly = poly5.fit_transform(X)
poly5.fit(X_poly,Y)
lin_poly5_reg = LinearRegression()
lin_poly5_reg.fit(X_poly,Y)

plt.scatter(X, Y, color = 'blue')
plt.plot(X, lin_poly5_reg.predict(X_poly), color = 'red')
plt.xlabel('Age')
plt.ylabel('Salary')

## Using 2nd degree polynomial

poly2 = PolynomialFeatures(degree = 2)
X_poly = poly2.fit_transform(X)
poly2.fit(X_poly,Y)
lin_poly2_reg = LinearRegression()
lin_poly2_reg.fit(X_poly,Y)

plt.scatter(X, Y, color = 'blue')
plt.plot(X, lin_poly2_reg.predict(X_poly), color = 'red')
plt.xlabel('Age')
plt.ylabel('Salary')

#OUTPUT