00:02
Okay, so we're going to look at a regression output that is pre -given to us for an application relating to a maintenance expense, so let's say dollars per month, and the usage, so hours per week for a particular brand of computers.
00:21
So what we're going to do here is look at an annova table that is given to us, write an estimated regression equation based off of that.
00:29
Then using a t test, determine if that monthly expense, which again, we're looking at dollars per month, is related to the actual usage hours per week of that computer.
00:40
And we're actually going to look at a level of significance of 0 .05.
00:44
And then at the end, to wrap it up, we're going to look at whether or not that estimate a regression equation is actually going to be a good fit or not based on what we're solving for.
00:52
So what we have here, which again is given to you, so no need to solve, is a regression output with the degrees of freedom, 189, and then you have your residual and your total.
01:12
And then some of squares here would be 1575 .76, 349 .14, and then 1924 .4 .4 .4 .4.
01:24
And then 1924 .4.
01:26
0 .90 and then they've left everything else blank for this problem so we're going to go down here and again this is going to be already pre -given to you but let's say for some reason you would have to find some of this what you would do is go into excel and if you are using you are using a microsoft windows product or like a dell or anything like that you would go to the upper right -hand corner there should be a data analysis toolkit you go into that you press that you would go to the drop down find regression and then it will prompt you for your x and y variables as long as you have those inputted as a table you should be able to highlight and it will give you an output that similar to this so just in case you don't need it right now but for feature reference that is how you would do it for a mac i think that's as simple as you can just kind of go on and and look um and see what the that would look like for you so just in terms of space, i'm just going to abbreviate.
02:35
So your coefficient for your intercept should be that 6 .1092.
02:42
Your usage, 0 .8951.
02:47
Your standard error, excuse me, that should be error instead of deviation.
02:54
So your standard error is going to be 0 .9361.
03:01
0 .149.
03:03
And then t -stap, we're going to look at later, and then a p value.
03:08
But they've left that blank.
03:10
So here's all the given information that we have to be able to solve.
03:13
Now, based off of all this information, what we want to do is look and see what our regression equation would be.
03:22
So what we would do is, again, we're prompted to look at what is the correlation relating between a maintenance expense and the usage.
03:33
So what you would do here is look at your coefficients, which is right here, to be able to create your equation.
03:41
So yhap would equal to the intercept coefficient of 6 .1092 plus your usage, 0 .8951x.
03:54
Because again, for every time or the usage that you're having, so hours per week, will change the maintenance expense or your intercept.
04:03
On a monthly scale.
04:05
So you need to make sure those are the variables that could change.
04:09
So your x and your y.
04:10
So that would be your estimated regression equation.
04:14
So that is what we're going to go off of.
04:18
But we need to determine whether or not there is a relationship between that maintenance expense and the actual usage.
04:25
And again, as mentioned, we're actually going to look at 0 .5 level of significance right now.
04:30
So what do we need to do for this? we need to create a null an alternative hypothesis to be able to test this.
04:38
So it would look like this.
04:40
H of o, and then we would have b sub 1, 0, and h of 1 b1 does not equal to 0.
04:57
So in this scenario, we have b1 is going to be our constant...