What is noise in regression?

#1

Sir, in the tutorial what does the term noise justify in the expression
y=f(x)+c+noise
what could be a noise in data manipulation?

#2

Hey @Kunnaal, i would like to explain you this by an example,

x y
1 1
2 2
3 3
4 4
5 5

Now if you try to fit this in linear regression model, you will be able to find a perfect fit line y = x and every data point perfectly satisfies this. So noise = 0

x y
1 1
2 2.1
3 2.9
4 4
5 5

Now if you want to fit again now also the line y = x come but now 2 and 3 point didn’t fit the game, and this means they contains some noise, their values should be 2 and 3. So this is what referred as noise in the dataset.

Hope this resolved your doubt.
Plz mark the doubt as resolved in my doubts section :blush:

#3

thank you
that was helpful