Would you like to inspect the original subtitles? These are the user uploaded subtitles that are being translated:
1
00:00:00,000 --> 00:00:05,002
(upbeat music)
2
00:00:05,002 --> 00:00:07,000
- [Instructor] I hope you've had a chance to work
3
00:00:07,000 --> 00:00:09,009
with the problem presented in this workbook.
4
00:00:09,009 --> 00:00:12,009
I'd like to now show you how I would solve it.
5
00:00:12,009 --> 00:00:15,009
And again, my sample file is Chapter_5_Challenge
6
00:00:15,009 --> 00:00:17,008
and you can find it in the chapter 5 folder
7
00:00:17,008 --> 00:00:21,004
of the exercise files collection.
8
00:00:21,004 --> 00:00:24,006
The first task is to go to cell F 2
9
00:00:24,006 --> 00:00:26,004
and calculate the co-variance
10
00:00:26,004 --> 00:00:30,003
between the values and the customers and sales columns.
11
00:00:30,003 --> 00:00:32,007
So in F 2 I'll type in equal sign
12
00:00:32,007 --> 00:00:36,002
and I will start typing covar
13
00:00:36,002 --> 00:00:38,004
and because you can never be sure
14
00:00:38,004 --> 00:00:40,003
that you have all of your data,
15
00:00:40,003 --> 00:00:42,005
I'm going to use COVARIANCE.S
16
00:00:42,005 --> 00:00:45,002
which assumes that we have a sample,
17
00:00:45,002 --> 00:00:46,006
so I'll highlight that, press tab,
18
00:00:46,006 --> 00:00:49,006
and now I need to identify the two arrays.
19
00:00:49,006 --> 00:00:51,009
The first one is customers
20
00:00:51,009 --> 00:00:56,002
and those values are in A 2 through A 12
21
00:00:56,002 --> 00:01:00,006
then a comma and then sales, which are in column B.
22
00:01:00,006 --> 00:01:04,006
So B 2 to B 12 I've selected there, right parenthesis
23
00:01:04,006 --> 00:01:11,003
and enter, and we get a covariance of 4023.16.
24
00:01:11,003 --> 00:01:13,002
How you interpret that covariance value
25
00:01:13,002 --> 00:01:15,002
depends on your original data.
26
00:01:15,002 --> 00:01:18,009
In this case, it does appear just from visual inspection
27
00:01:18,009 --> 00:01:20,002
that the number of customers
28
00:01:20,002 --> 00:01:22,006
in sales tend to co-vary together
29
00:01:22,006 --> 00:01:27,003
and 4,023 is about twice the maximum value
30
00:01:27,003 --> 00:01:29,008
in the sales column, so that's a good indicator
31
00:01:29,008 --> 00:01:32,007
that there probably is a relationship.
32
00:01:32,007 --> 00:01:34,008
Now let's do a correlation
33
00:01:34,008 --> 00:01:38,006
for the same two data sets and see what that does.
34
00:01:38,006 --> 00:01:41,006
So I'll go to cell F 4, type in equal sign,
35
00:01:41,006 --> 00:01:45,009
and we'll have CORREL, C O R R E L,
36
00:01:45,009 --> 00:01:48,006
and again, array1 and array2,
37
00:01:48,006 --> 00:01:52,006
array1 is A 2 through A 12, then a comma,
38
00:01:52,006 --> 00:01:58,005
and then array2, B 2 to B 12, right parenthesis, and enter.
39
00:01:58,005 --> 00:02:02,004
And I get a correlation of customers and sales of 0.84
40
00:02:02,004 --> 00:02:04,005
and that is very high.
41
00:02:04,005 --> 00:02:08,001
Remember that correlation varies between negative one
42
00:02:08,001 --> 00:02:10,007
and one with a value of one
43
00:02:10,007 --> 00:02:13,005
meaning that the values are perfectly correlated.
44
00:02:13,005 --> 00:02:16,006
And in a case where you have a value of 0.84
45
00:02:16,006 --> 00:02:20,007
with only 11 data points, that's a very good indication
46
00:02:20,007 --> 00:02:25,006
that there is a correlation between those two sets of data.
47
00:02:25,006 --> 00:02:26,009
Finally, let's go ahead
48
00:02:26,009 --> 00:02:31,008
and create a grid that calculates the correlation
49
00:02:31,008 --> 00:02:34,007
for each pair of columns.
50
00:02:34,007 --> 00:02:38,008
So I'll start by clicking and cell G 8
51
00:02:38,008 --> 00:02:41,008
and I want to create a formula that will calculate
52
00:02:41,008 --> 00:02:44,007
the correlation of the customer's column,
53
00:02:44,007 --> 00:02:48,006
column A with itself and I'll be able to copy that formula
54
00:02:48,006 --> 00:02:50,006
throughout the rest of the grid
55
00:02:50,006 --> 00:02:54,008
and find correlations between every pair of columns.
56
00:02:54,008 --> 00:03:00,003
So in G 8 I'll type = and then correl, C O R R E L,
57
00:03:00,003 --> 00:03:06,004
left parentheses and array1 is A 2 through A 12.
58
00:03:06,004 --> 00:03:09,001
However, with the way I'm creating my worksheet,
59
00:03:09,001 --> 00:03:10,007
I don't want that reference to change
60
00:03:10,007 --> 00:03:12,002
when I copy the formula.
61
00:03:12,002 --> 00:03:14,009
So I'll press F4
62
00:03:14,009 --> 00:03:20,007
and that makes the entire reference absolute and unchanging,
63
00:03:20,007 --> 00:03:24,000
type a comma and then I'll select A 2
64
00:03:24,000 --> 00:03:28,002
through A 12 again, right parenthesis and enter,
65
00:03:28,002 --> 00:03:29,009
and I get a correlation of one
66
00:03:29,009 --> 00:03:32,002
which as I mentioned earlier is correct.
67
00:03:32,002 --> 00:03:35,003
I'm comparing customers to itself.
68
00:03:35,003 --> 00:03:40,004
Now I will double click cell G 8 and copy the formula,
69
00:03:40,004 --> 00:03:44,005
so remember I have copied the cell's formula text.
70
00:03:44,005 --> 00:03:46,006
I have not actually copied the cell.
71
00:03:46,006 --> 00:03:47,009
So it is important
72
00:03:47,009 --> 00:03:51,003
that you not just click a cell and press control C.
73
00:03:51,003 --> 00:03:53,000
So I've already copied the values
74
00:03:53,000 --> 00:03:56,003
and then press escape to stop editing the cell,
75
00:03:56,003 --> 00:04:00,005
double click cell G 9, press control V to paste the text,
76
00:04:00,005 --> 00:04:03,009
and now, because I want to change the absolute references
77
00:04:03,009 --> 00:04:07,007
from A 2 to A 12, to B 2 to B 12
78
00:04:07,007 --> 00:04:13,008
I will edit the first range reference so I have B 2 to B 12
79
00:04:13,008 --> 00:04:16,001
and because I'm comparing customers sales
80
00:04:16,001 --> 00:04:18,000
I will leave A 2 to A 12
81
00:04:18,000 --> 00:04:22,002
as the relative references unchanged, press enter.
82
00:04:22,002 --> 00:04:26,000
And I get a correlation of 0.842
83
00:04:26,000 --> 00:04:29,002
which we can see is correct because it matches
84
00:04:29,002 --> 00:04:33,006
the calculation we did earlier in cell F 4.
85
00:04:33,006 --> 00:04:35,009
Now I will double click cell G 10,
86
00:04:35,009 --> 00:04:38,007
press control V, there's my formula again
87
00:04:38,007 --> 00:04:42,008
and I will do the same thing for the first array
88
00:04:42,008 --> 00:04:47,007
except this time I'm changing it to column C returns.
89
00:04:47,007 --> 00:04:53,000
So C 2 to C 12, comparing it to A 2 to A 12, press enter,
90
00:04:53,000 --> 00:04:56,003
and I see a very slightly negative correlation
91
00:04:56,003 --> 00:04:59,001
but that is very close to zero.
92
00:04:59,001 --> 00:05:01,007
Now I can copy the results
93
00:05:01,007 --> 00:05:06,001
of these three formulas over to cover columns H and I,
94
00:05:06,001 --> 00:05:10,001
so I will select G 8 through G 10
95
00:05:10,001 --> 00:05:11,002
and drag the fill handle
96
00:05:11,002 --> 00:05:16,002
to the right until it covers the same rows in column I,
97
00:05:16,002 --> 00:05:18,000
and there I get the values
98
00:05:18,000 --> 00:05:20,002
and it looks like everything is correct
99
00:05:20,002 --> 00:05:23,007
because I have customers and customers at one,
100
00:05:23,007 --> 00:05:27,006
sales and sales at one, and returns and returns at one.
101
00:05:27,006 --> 00:05:31,005
And I can see also that sales and returns do not
102
00:05:31,005 --> 00:05:33,000
appear to be correlated,
103
00:05:33,000 --> 00:05:38,008
a correlation value of 0.126 let's call it, is very small.
104
00:05:38,008 --> 00:05:41,004
So it appears that the only meaningful relationship
105
00:05:41,004 --> 00:05:43,005
at least from a correlation standpoint
106
00:05:43,005 --> 00:05:48,000
that I have is between sales and customers per day.
8478
Can't find what you're looking for?
Get subtitles in any language from opensubtitles.com, and translate them here.