linear regression

[scarlet tiger]

When we say "we keep moving this line through the data points," we mean that we are trying to find the line that best represents the relationship between the variables by adjusting the slope and intercept of the line. We do this by measuring the square distance between the data points and the regression line, and trying to minimize this distance.

[crystal wadi]

You try to minimise the sum of the squared distances

Idk how good you are at maths, but mathematically (for one independent variable x) you want to find

[fading fern]

@tranquil elm

im_maro Uploaded Some Code

Uploaded these files to a Gist

1000_Companies.csv

View The Gist

For safety reasons we do not allow file attachments.

[crystal wadi]

That's what you're trying to solve for a linear regression (with one independent variable)

In simple terms, you want to find the slope and bias of the regression line that minimises the sum of the squared residuals

Where the sum of squared residuals is the sum of the distances of each data point to the regression line squared

Same concept

[crystal wadi]

The right hand side the mx_i+c is the regression line you try

And to mae predictions using your regression line, you plug in each data point into the equation

Then you find the squared residuals like that

c is not the coefficient of y but the y-intercept

or the bias

but yeah, conceptually that's correct

[crystal wadi]

I feel like you don't quite understand the concept behind linear regression

https://www.youtube.com/watch?v=PaFPbb66DxQ