00:01
We have these four data points, 04 -13 -1120, and we want to find the line that best goes through those data points, the best fits those data points, so our linear model.
00:17
So we construct our sum of squared errors.
00:21
This data point gives us b minus 4 squared.
00:25
This one gives us a plus b minus 3 squared.
00:27
This one gives us a plus b minus 1 squared.
00:31
And this gives us 2a plus b squared.
00:37
So we want to minimize this.
00:40
So we can take the partials, partial derivatives, but the a and b, and set them equal to 0.
00:46
And let's see here.
00:48
We get, i didn't go through all the details here, but i'm assuming that you can do that.
00:54
The partial with respect to a is 12a plus 8b minus 8, and that has to be 0.
01:01
And then the partial receptor b is 8 times a plus b minus 2.
01:06
That has to be equal to 0...