WEBVTT

1
00:00:00.050 --> 00:00:01.080
- [Narrator] This is gonna be an interesting

2
00:00:01.080 --> 00:00:03.060
milestone for us 'cause we're gonna have

3
00:00:03.060 --> 00:00:07.020
a lot of pieces come together in a case study.

4
00:00:07.020 --> 00:00:10.030
In this video, we're gonna be setting up

5
00:00:10.030 --> 00:00:13.000
an example of simultaneous regression.

6
00:00:13.000 --> 00:00:17.030
Let's go to Analyze, Regression, Linear.

7
00:00:17.030 --> 00:00:19.090
We're gonna have waste tons be the dependent,

8
00:00:19.090 --> 00:00:24.020
and all of the available scale variables,

9
00:00:24.020 --> 00:00:27.030
all five of them will be independent.

10
00:00:27.030 --> 00:00:30.040
Now, one of the features that makes this simultaneous,

11
00:00:30.040 --> 00:00:32.050
the primary one in fact,

12
00:00:32.050 --> 00:00:36.020
is the fact that method is going to be Enter,

13
00:00:36.020 --> 00:00:39.080
and we are not gonna be utilizing the Next button.

14
00:00:39.080 --> 00:00:41.060
Those two things taken together

15
00:00:41.060 --> 00:00:43.090
is what really makes this simultaneous.

16
00:00:43.090 --> 00:00:45.090
So since we are approaching this

17
00:00:45.090 --> 00:00:47.090
as a simultaneous regression,

18
00:00:47.090 --> 00:00:50.080
let's talk about what settings are appropriate for that.

19
00:00:50.080 --> 00:00:54.060
We're gonna click on the Statistics sub-menu.

20
00:00:54.060 --> 00:00:57.050
And we don't need R squared changed,

21
00:00:57.050 --> 00:00:59.030
'cause that's really a hierarchical

22
00:00:59.030 --> 00:01:00.050
regression kind of thing.

23
00:01:00.050 --> 00:01:02.070
But we are gonna request descriptives.

24
00:01:02.070 --> 00:01:05.060
We are interested in partial correlations.

25
00:01:05.060 --> 00:01:09.050
And we're not gonna request collinearity diagnostics now.

26
00:01:09.050 --> 00:01:11.060
We're gonna address that later.

27
00:01:11.060 --> 00:01:15.050
Also, we're not gonna request a Durbin-Watson statistic.

28
00:01:15.050 --> 00:01:18.040
The assumption is that we've done a good check

29
00:01:18.040 --> 00:01:21.000
of multiple regression assumptions

30
00:01:21.000 --> 00:01:23.060
prior to getting to this stage.

31
00:01:23.060 --> 00:01:26.040
Also remember, it's always terribly important

32
00:01:26.040 --> 00:01:29.060
to explore your data and check all your assumptions

33
00:01:29.060 --> 00:01:31.090
before you get to this point.

34
00:01:31.090 --> 00:01:34.090
What we will do, however, is request confidence intervals

35
00:01:34.090 --> 00:01:39.040
around our data coefficients, and then click on Continue.

36
00:01:39.040 --> 00:01:43.040
Moving on, we're gonna go ahead and request Residuals Plots.

37
00:01:43.040 --> 00:01:46.020
So recall that the standard way to do this

38
00:01:46.020 --> 00:01:51.070
is ZRESID and the Y and ZPRED in the X.

39
00:01:51.070 --> 00:01:54.060
We're also gonna request a histogram.

40
00:01:54.060 --> 00:01:58.060
And since we now have multiple independent variables,

41
00:01:58.060 --> 00:02:05.000
we're gonna request produce all partial plots.

42
00:02:05.000 --> 00:02:08.040
Under Save we have numerous options,

43
00:02:08.040 --> 00:02:12.090
including ways of diagnosing outliers,

44
00:02:12.090 --> 00:02:14.010
like Cook's distance,

45
00:02:14.010 --> 00:02:17.010
which is perhaps the most famous check for outliers.

46
00:02:17.010 --> 00:02:19.090
This, again, is gonna be addressed at a different time.

47
00:02:19.090 --> 00:02:23.050
We will not request that now.

48
00:02:23.050 --> 00:02:27.030
Finally, under Options, we have a number of options

49
00:02:27.030 --> 00:02:29.060
for stepping method, and so on.

50
00:02:29.060 --> 00:02:33.070
These are appropriate only when doing stepwise.

51
00:02:33.070 --> 00:02:36.010
So we also don't have to worry about them.

52
00:02:36.010 --> 00:02:37.010
So we're ready to go.

53
00:02:37.010 --> 00:02:38.060
We're gonna click on OK.

54
00:02:38.060 --> 00:02:40.070
And in the next video, we're gonna be

55
00:02:40.070 --> 00:02:40.070
exploring the output in detail.