Page 251 - Touhpad Ai

120
100

80

Marks (Y) 60 residual (e)=Observed value -- Predicted Value

40
20

0
0 1 2 3 4 5 6 7 8 9 10
No. of Hours Studied (X)

Regression-How good is the Line?

u Linear regression aims to find the best-fitting straight line through the points.
u If data points are closer to the line of best fit (less residual error), it means the correlation between the two variables
is higher. That means, the relationship between the two variables is strong.
u The regression line is also called ‘Line of Least Squares of Errors’ because the lower the residual errors, the better.
u Each data point has one residual.
u Outliers are data points that are significantly different from other data points. Outliers can significantly affect the slope
and intercept, skewing the best-fit line.

21 st
VIDEO SESSION Century #Digital Literacy
Skills
Scan the QR code or visit the following link to watch the video:
Linear Regression Algorithm
https://www.youtube.com/watch?v=E5RjzSK0fvY&t=527s
After watching the video, answer the following question:
What do you learn from the video?

BRAINY The least squares regression method was first published by mathematicians Legendre in
FACT 1805 and Carl Friedrich Gauss in 1809. Both used linear regression to predict the movement of
planets around the sun. Gauss later published an improved method in 1821.

Finding the Line of Best Fit
A child’s age and height are recorded for study. Let’s find the line of best fit:
Age (years) Height (cm)
0 20
1 40
2 60
3 80
4 92
5 105

Data Modelling and Simple Linear Regression 249