1
00:00:01,350 --> 00:00:08,130
In the last video, we have seen the result of our linear regression model, in this lecture, we will

2
00:00:08,130 --> 00:00:12,840
learn how to use that result to predict the future values.

3
00:00:13,950 --> 00:00:15,240
Now, if you remember.

4
00:00:17,090 --> 00:00:19,640
This is the equation of linear regression.

5
00:00:21,520 --> 00:00:30,640
Y equal to be not let's be the next one plus B two x2 plus B three 3x three up to be a Nexen, this

6
00:00:30,640 --> 00:00:31,610
value is over.

7
00:00:32,290 --> 00:00:40,270
That is the intercept value we have calculated B one, B to be three up to be in and these values are

8
00:00:40,270 --> 00:00:41,190
located here.

9
00:00:41,620 --> 00:00:45,780
So this one is one, this one is B2, this one is B three.

10
00:00:46,120 --> 00:00:48,760
This one is before and so on.

11
00:00:50,780 --> 00:00:58,700
No, afeared be not blessed be the next one that is this BE1 into this one.

12
00:00:59,460 --> 00:01:06,530
And similarly, if we are the multiplication of these B with other independent values, we will be able

13
00:01:06,530 --> 00:01:08,620
to find out the valuation of our business.

14
00:01:08,630 --> 00:01:10,010
That is the variable.

15
00:01:11,790 --> 00:01:17,100
So using this equation, we can find out the predicted values of why.

16
00:01:18,360 --> 00:01:22,860
So let's copy this coefficient values in our previous shoot.

17
00:01:27,970 --> 00:01:30,610
No, I have copied the Confucian values.

18
00:01:30,640 --> 00:01:31,930
This one is Betelnut.

19
00:01:33,440 --> 00:01:43,410
I can see that and for the so large, the value is 469 and you can see just about the store large.

20
00:01:43,670 --> 00:01:51,660
We have Mr. Doko vision value for this, very similarly for a store equal to medium value.

21
00:01:51,710 --> 00:01:58,610
We are getting is a 32 and you can see that here also for his store medium that we would lose 800 or

22
00:01:59,570 --> 00:01:59,820
so.

23
00:01:59,840 --> 00:02:06,260
VFP said the conversion values corresponding to this variables on the top row of this shoot.

24
00:02:08,030 --> 00:02:17,150
So now if we take the sum of bitterness, plus with our next one, we take two weeks to be three x three

25
00:02:17,630 --> 00:02:21,920
to four x four and so on, we will be able to get the valuation value.

26
00:02:23,290 --> 00:02:27,040
So now let's predict the values for this first store.

27
00:02:28,290 --> 00:02:35,540
Now, the equation is this, so we'll follow this equation first, we have to add we betelnut.

28
00:02:36,120 --> 00:02:40,410
So our first term is Betelnut, which is this.

29
00:02:41,850 --> 00:02:48,960
And then the second term is the multiplication of this value coefficient and to.

30
00:02:50,550 --> 00:02:58,500
This X1, there's the value of forex, rebel or independent or rebel, then we have the application

31
00:02:58,500 --> 00:03:00,270
of this into this.

32
00:03:01,800 --> 00:03:03,930
Plus, this.

33
00:03:05,420 --> 00:03:06,470
And to this.

34
00:03:07,760 --> 00:03:09,260
There's this.

35
00:03:10,410 --> 00:03:16,390
And to this, so this is the Corporation for Sumitra and this is the variable city metro.

36
00:03:16,410 --> 00:03:19,200
So we are multiplying both of these two values.

37
00:03:20,140 --> 00:03:29,710
Then we have coefficient of investment and to the value of investment, plus coefficient of number of

38
00:03:29,710 --> 00:03:32,560
competitors and the number of competitors.

39
00:03:34,890 --> 00:03:35,460
Plus.

40
00:03:36,770 --> 00:03:37,400
This.

41
00:03:38,510 --> 00:03:39,290
And to.

42
00:03:40,340 --> 00:03:40,850
This.

43
00:03:41,800 --> 00:03:42,400
Plus.

44
00:03:43,670 --> 00:03:46,310
This to this.

45
00:03:47,930 --> 00:03:55,490
So the value we will be getting from this equation will be the predicted value of valuation for this

46
00:03:55,490 --> 00:03:55,820
system.

47
00:03:56,300 --> 00:04:06,200
So the predicted value for this store one was 86000 and in reality, we are getting the valuation this

48
00:04:06,200 --> 00:04:06,890
71.

49
00:04:08,260 --> 00:04:12,520
And similarly, we can find appropriate value for the rest of those stoats.

50
00:04:16,460 --> 00:04:22,190
And we can expand this formula, but before expanding, we have to fix the cell referencing.

51
00:04:23,790 --> 00:04:31,110
So just fix the cell referencing for all this coefficients, we don't need to fix the cell referencing

52
00:04:31,110 --> 00:04:39,180
for this X variable because when we are dragging down, we want these values to be changed accordingly.

53
00:04:40,550 --> 00:04:41,930
So BE2.

54
00:04:43,040 --> 00:04:45,750
Is our goal for you, it is also efficient.

55
00:04:45,770 --> 00:04:52,280
I have to is also efficient due to its coefficient, etc. is also vision.

56
00:04:53,660 --> 00:04:56,060
I talked to two.

57
00:04:59,370 --> 00:05:01,420
Care to find to.

58
00:05:04,610 --> 00:05:10,790
So we have extra fencing for all the coefficients, no, let's try dragging it down.

59
00:05:11,980 --> 00:05:19,060
Now, for example, if you just click, you can see that the formula is working fine, we are adding

60
00:05:19,060 --> 00:05:25,930
Binet with the multiplication of between X1 better to extra Pittodrie extra and so on.

61
00:05:26,460 --> 00:05:27,960
So the formula is working fine.

62
00:05:31,130 --> 00:05:36,870
And you can see that for the second they saw the predicted value is sixty five thousand.

63
00:05:37,040 --> 00:05:40,050
And in reality, we are getting a valuation of 68.

64
00:05:40,430 --> 00:05:44,900
Similarly, you can compare these values for other stores as well.

65
00:05:45,610 --> 00:05:52,490
Let's expand the line so that we can get this value for all the 13000 stores.

66
00:05:54,080 --> 00:05:56,880
Now, how are we going to use this model?

67
00:05:57,500 --> 00:06:03,380
So suppose if your company or SBA retailers want to open another store.

68
00:06:04,590 --> 00:06:08,910
And they want to protect the valuation before opening that store.

69
00:06:10,080 --> 00:06:14,970
So here you can fill out all the details and you will get the predicted value of that.

70
00:06:15,040 --> 00:06:15,360
So.

71
00:06:17,100 --> 00:06:19,320
Let's fix the Penn.

72
00:06:21,240 --> 00:06:27,760
The shortcut is w f or else you can just select the cell, go to this view option.

73
00:06:28,890 --> 00:06:36,090
And click on this free speech on what this will do is if you just scroll down, this top three rows

74
00:06:36,390 --> 00:06:39,230
are not going to move out of your view.

75
00:06:40,290 --> 00:06:44,820
And in a way, you will be able to see the hiders of your data.

76
00:06:46,490 --> 00:06:48,080
While scrolling as well.

77
00:06:49,900 --> 00:06:52,860
So, yeah, let's suppose we are opening a new store.

78
00:06:59,700 --> 00:07:05,880
And you have some data of your restored, so the first information that you have about this story is

79
00:07:05,880 --> 00:07:09,830
that it is going to be a medium size store.

80
00:07:10,020 --> 00:07:11,010
So just straight.

81
00:07:11,010 --> 00:07:11,490
Medium.

82
00:07:15,660 --> 00:07:19,130
Now for location, location is going to be a commercial location.

83
00:07:26,120 --> 00:07:27,860
Now, the city is Michael.

84
00:07:29,570 --> 00:07:30,740
So just fill out.

85
00:07:32,160 --> 00:07:36,390
The information of the new search that you are planning to open.

86
00:07:37,440 --> 00:07:46,230
No investment, the company is planning to invest around 55000 for this store, the number of competitors

87
00:07:46,350 --> 00:07:48,240
in this area is just too.

88
00:07:49,240 --> 00:07:52,600
And the population is around 12000.

89
00:07:55,120 --> 00:07:58,270
The average household income is twenty five thousand.

90
00:08:03,950 --> 00:08:10,520
So with the data that we have entered, we are getting a valuation predicted valuation of 90000.

91
00:08:11,270 --> 00:08:16,070
So we are going to invest 55000 and we are getting a valuation of 90000.

92
00:08:16,370 --> 00:08:20,510
So I think this is good for S.T. to open this location.

93
00:08:22,140 --> 00:08:30,150
So similarly, if Steve wants to open some new store, they just have to put information of the store

94
00:08:30,150 --> 00:08:35,430
that they are going to open and they will be able to find the valuation with this linear regression

95
00:08:35,430 --> 00:08:35,780
model.

96
00:08:37,500 --> 00:08:43,740
So that's how we protect the values of our dependent variable using the linear regression.

97
00:08:46,660 --> 00:08:48,630
So that's all for this case study.

98
00:08:49,120 --> 00:08:49,660
Thank you.