00:01
So for this problem, the first thing that we're going to need to do is to calculate the standard error of the estimate, which is equal to the square root of the sum of the squared difference between each observed y value and the corresponding predicted value.
00:20
So we do y minus y hat i squared divided by n minus two, where we're told that our regression equation, oh, actually, no, never mind.
00:31
We're told that se, we're actually given that, so that makes our life a little bit easier.
00:38
We're told that se is equal to 1 .127.
00:42
Once we have the standard error of the estimate, we'd want to find the margin of error, which is equal to the critical t value for our level of confidence, which i'll get to in a second, times the standard error of the estimate, times the square root of 1 plus 1 over n plus n times x not, where x not is the value, the x value that we'll be plugging in, that we'll be plugging in, minus x bar squared divided by n times the sum of x squared, so that's the sum of the squared x values, minus the square of the sum of x.
01:29
We're taking the square root of all of that.
01:32
So i'll note here that that critical t value tc is going to be t for n minus 2 degrees of freedom and a tail proportion of 1 minus alpha over 2.
01:43
So in this case that's a tail proportion of 0 .025.
01:47
Now i'll bring up a t distribution table...