1
00:00:00,880 --> 00:00:03,940
In this video, we will learn about variable transformation.

2
00:00:05,730 --> 00:00:13,980
As you can see in our dataset, we have four distance variables, these distance variables are giving

3
00:00:13,980 --> 00:00:19,010
us the distance of the house from four different employment hubs.

4
00:00:20,610 --> 00:00:27,390
Now keeping for distance variables in our dataset is over, representing the importance of the employment

5
00:00:27,390 --> 00:00:27,840
variable.

6
00:00:29,290 --> 00:00:36,700
We need to transform these four variables into one single distance variable, which can represent the

7
00:00:36,700 --> 00:00:39,040
employment opportunity near that house.

8
00:00:40,710 --> 00:00:47,070
There can be several ways to transform these four into one, either we can choose the minimum of the

9
00:00:47,070 --> 00:00:54,810
distance or we can choose the maximum of these four distances, or we can just take an average of these

10
00:00:54,810 --> 00:00:55,620
four distances.

11
00:00:57,300 --> 00:01:03,900
Usually we have to take into account the business scenario in which we are to decide which transformation

12
00:01:03,900 --> 00:01:05,370
matter we are going to use.

13
00:01:07,140 --> 00:01:13,620
Here, I'll use the mean method, that is, I'll replace all these four variables by the mean of these

14
00:01:13,620 --> 00:01:15,030
four variables.

15
00:01:16,990 --> 00:01:20,670
Depending on your business scenario, you have to take the decision for your problem.

16
00:01:22,510 --> 00:01:25,240
So I'll add a new variable.

17
00:01:26,580 --> 00:01:33,870
To add a new I need a new column, so I write on the column name and I inserted a new column and I'm

18
00:01:33,870 --> 00:01:37,800
naming it average based EVGA and this could be used.

19
00:01:40,360 --> 00:01:48,460
Now, in this variable, I'll put the average value of the other four variables to this is average,

20
00:01:50,200 --> 00:01:54,070
what these four variables render.

21
00:01:55,900 --> 00:02:03,700
Now, to extend this formula for other cells, you can just double click on this bottom right corner

22
00:02:03,700 --> 00:02:09,330
of the cell and this formula will be extended to other cells below the cell.

23
00:02:12,110 --> 00:02:18,980
So this value of 4.0 aid is the average of these four distances, four point three five, three point

24
00:02:18,980 --> 00:02:21,260
eight one or one minute and 4.0 one.

25
00:02:22,720 --> 00:02:28,140
Similarly, the average distance in the next observation is the average of next for distant variables.

26
00:02:30,950 --> 00:02:36,460
Now that we have the average distance variable, we do not need the other four distance variables,

27
00:02:37,340 --> 00:02:40,950
so therefore we will be deleting those four distant variables.

28
00:02:41,840 --> 00:02:48,860
But before we delete these four variables, we first need to change the value in this cell from this

29
00:02:48,860 --> 00:02:51,240
formula to the actual value.

30
00:02:52,430 --> 00:02:55,070
So we will select the values in all these cells.

31
00:02:56,580 --> 00:02:58,580
Copied them by pressing control, see?

32
00:02:59,980 --> 00:03:04,930
Then right, click on the cell and paste these values.

33
00:03:06,010 --> 00:03:12,720
Now we have the exact same value in this cell, but instead of the formula, we have the value here.

34
00:03:13,960 --> 00:03:17,470
Now, if I delete these four columns, it will not impact.

35
00:03:19,430 --> 00:03:26,900
Devaluing the average tenth column to have selected those four columns and clicked on the lead option.

36
00:03:27,920 --> 00:03:37,460
So this is how we identified variables which need transformation, we transform them into a new variable

37
00:03:37,910 --> 00:03:40,040
and we deleted the old variables.