Question

Hey, I'm working on Scatter plots and correlation and I wanted to know what does, "find the least-square regression line?" What does that mean?

Answer 1

oh i did a whole paper on this

Answer 2

would u like me to send it to you

Answer 3

it shows everything step by step

Answer 4

Yeah, that would be great. Thanks

Answer 5

ok ill attach it here
give me a minute

Answer 6

Alright 😊

Answer 7

A least squares regression line is a fitted line so that the squared distance between a point on the line (called a residual) and its corresponding observed point is a minimum.

Answer 8

I didn't instant understand any of that

Answer 9

Not*

Answer 10

It's a bit complicated but basically you sometimes draw lines of best fits through a scatterplot just to indicate a linear trend, right?

Answer 11

now in high school a lot of this is drawn 'by sight'.
So if you asked 5 different students to draw that fitted line by eye, they'd come up with slightly different answers.

Thus statistics has developed ways to map a fitted line (one of which is called a least squares regression line) to a scatterplot so that everyone will get the same line and even better, an equation for that line.

Answer 12

To do this they simply measure the vertical distances from the line to the actually point (x,y)



The line that makes the sum of these distances (known as residuals) when they are squared, the smallest is known as the least squares regression line and is considered the best fit to the points. There is a formula to calculate this line involving the standard deviation of the x values, the stdev of the y values, the correlation coefficient, r and the mean of x and mean of y:

http://www.stat.wmich.edu/s216/book/node126.html