00:01
We're looking at a regression model here.
00:04
A nice simple linear regression model.
00:07
And the data was gathered for two variables, a and b, and the correlation coefficient of r was calculated to be this, negative 0 .89.
00:16
So which of these are true? this is true.
00:21
The linear regression line provides a strong fit to the observed data.
00:24
Yes, because the correlation coefficient goes between negative 1 and 1.
00:31
And the closer it is to 1 or negative 1, the stronger it is.
00:34
And this is close to negative 1, so we'd say it's strong.
00:38
Beta not is known as the sample intercept.
00:41
It's the intercept, but no, it's this beta not is the population intercept.
00:47
The sample intercept would be something like this, beta 1.
00:51
Variable a is known as the explanatory variable.
00:54
B is the response.
00:55
No, it's the reverse.
00:56
B is the thing that explains a.
01:01
The proportion of variability in a, that's explained by the regression model, is equal to 79 .21%.
01:08
Yes, that is true.
01:10
And you get that by squaring the r value.
01:13
You take the correlation coefficient squared, and that is, it's called the coefficient of determination, which is what r squared is, and that's what r squared is.
01:22
Negative 0 .9 squared is, it's 0 .7921, but as a percent, it's 79 .21.
01:29
Variable b causes variable a.
01:32
The issue here is the word causes.
01:35
We're not looking for, we can never say something causation.
01:38
There's correlation causation here.
01:43
The graph of the linear relationship between a and b slopes up...