WEBVTT

1
00:00:00.000 --> 00:00:11.958
Here is the problem: Find the least-squares regression line for x equals 0, 1, 2, 3, 4 and y equals 1, 3, 7, 9, 11.

2
00:00:12.408 --> 00:00:22.683
Fit the straight line ŷ equals a plus bx that minimises the sum of squared vertical distances to the data, and measure how well it fits with r and r squared.

3
00:00:23.483 --> 00:00:24.969
Means of x and y.

4
00:00:25.419 --> 00:00:28.681
Sums of squares and cross-products about the means.

5
00:00:29.131 --> 00:00:29.688
Slope.

6
00:00:30.138 --> 00:00:33.575
Intercept: the line passes through the point of means.

7
00:00:34.375 --> 00:00:35.513
Regression line.

8
00:00:35.963 --> 00:00:47.074
Correlation and coefficient of determination. Strong positive linear association; about 98.26 percent of the variation in y is explained by the line.

9
00:00:47.874 --> 00:01:02.282
So the answer is ŷ equals 1 plus (13 over 5)x; r approximately 0.991241, r squared approximately 0.982558.

10
00:01:02.732 --> 00:01:12.264
The slope is the predicted change in y for each one-unit increase in x. Correlation measures the strength of the linear pattern, not whether x causes y.

11
00:01:13.064 --> 00:01:17.429
The residuals sum to 0, as they must for a least-squares line.