00:01
So for this problem, i'll note that there are two different methods.
00:05
There are two methods for hypothesis testing, a linear regression.
00:24
Method one would be basically answering the question, is the slope non -zero? or is the slope significantly non -zero, that is? and then the other way of testing a linear regression would be, is the relation significant? so for method 1, the appropriate set of hypotheses would be as follows.
01:12
The null hypothesis should be that the true value of the slope for our explanatory variable is equal to 0.
01:21
There's no relation as x changes y does not do anything.
01:28
The alternate hypothesis is that beta 1 simply does not equal 0.
01:34
So, of course, in here, the null hypothesis is that of no relation, and the alternate hypothesis is that of relation, effectively.
01:50
Then the other way, where we're focusing in directly on the significance of the relation, we have that the null hypothesis is that the population correlation coefficient, row, is equal to zero, and the alternate hypothesis is that the population correlation coefficient, does not equal zero...