Lab report least square fiiting line error ?

[jessicaw]

lab report least square fiiting line "error"?

what is least square fiiting line "error"?
y=A+Bx
It is said that there is errors involved in A and B;So the least square line has to be compared to some other lines to determine the errors. But how to do it? Are there any formulas??
Thank you!

 

[statdad]

what is least square fiiting line "error"?
y=A+Bx

You can google for "simple linear regression.

It is said that there is errors involved in A and B;

Yes, this is a statistical issue - but think of the errors as uncertainty

So the least square line has to be compared to some other lines to determine the errors.

No, you do not need to compare it to other lines. results on linear regression will explain that.

But how to do it? Are there any formulas??

Again, look at the linear regression stuff.

Thank you!

You are welcome. good luck. If you have more questions bring them back.

 

[jessicaw]

You can google for "simple linear regression.

Yes, this is a statistical issue - but think of the errors as uncertainty

No, you do not need to compare it to other lines. results on linear regression will explain that.

Again, look at the linear regression stuff.


You are welcome. good luck. If you have more questions bring them back.

Well the linear regression is too difficult. Can i rotate around the centroid of the least square line and find the maximum slope and minimum slope? But how to find the maximum slope is another problem.
You mention the error is uncertaincies earlier. Is it the uncertainty of points? Can i use it to calculate the uncertaincies(errors) of slope and y-intercept?