1
00:00:00,330 --> 00:00:04,650
Hello, all so welcome to this data analytics project for finance.

2
00:00:05,010 --> 00:00:12,390
So in this series, basically, I'm going to consider data of bank personal loan and here I'm going

3
00:00:12,390 --> 00:00:19,830
to perform some kind of analysis so that I can fetch some insights and some valuable information from

4
00:00:19,830 --> 00:00:21,750
this huge chunk of data.

5
00:00:22,020 --> 00:00:28,360
So this exactly is my data set on which I'm going to perform some sort of analysis over here.

6
00:00:28,680 --> 00:00:35,130
So very first, you have to just go across to the various basic steps of analysis.

7
00:00:35,130 --> 00:00:41,310
Very first, you have to read your data, then you have to go through with the people sitting stage

8
00:00:41,310 --> 00:00:46,560
of data where you are going to deal with the missing values and various other features as well.

9
00:00:46,830 --> 00:00:52,350
So I'm just going to open my own book, and this is exactly my very first.

10
00:00:52,350 --> 00:00:59,580
I have to import my some of the libraries for manipulation, for numerical computation, for visualizations,

11
00:00:59,580 --> 00:01:02,710
as well as for interactive results as well.

12
00:01:03,060 --> 00:01:08,760
So what I'm going to do is I'm just going to import my all the libraries over here so that if I have

13
00:01:08,760 --> 00:01:15,550
to import my panel, so I'm going to say I'm not as big and is basically all of us.

14
00:01:15,870 --> 00:01:18,300
So the second one is basically my name.

15
00:01:19,050 --> 00:01:24,000
So I'm going to import my name by and creating an alias as ENPI.

16
00:01:24,270 --> 00:01:34,620
And the third one is basically my basic library for vision purpose, which is my matplotlib and so on.

17
00:01:34,620 --> 00:01:40,320
For interactive graphs, you can import your seabourne for production.

18
00:01:40,320 --> 00:01:44,710
Even with the additions, you can import your blakley as well.

19
00:01:45,030 --> 00:01:47,790
So I'm just going to execute it very first.

20
00:01:47,800 --> 00:01:50,020
I have to read my data.

21
00:01:50,370 --> 00:01:54,920
So you will see this exactly is my actual data set.

22
00:01:54,930 --> 00:01:59,970
So I'm just going to copy this entire part because my data is somewhere available over here.

23
00:02:00,600 --> 00:02:05,960
So now I have to use a function known as Read an Excel.

24
00:02:05,970 --> 00:02:10,600
And in this basically very flat, I have to pass my entire.

25
00:02:10,860 --> 00:02:16,620
But then you can pass that over here to get to your dataset.

26
00:02:17,340 --> 00:02:20,990
So this is basically my actual data.

27
00:02:21,000 --> 00:02:27,970
And now you have to pass your what exactly the sheet name that you want to read.

28
00:02:28,350 --> 00:02:32,610
So if you are going to open this data, so this is exactly the data.

29
00:02:33,030 --> 00:02:39,850
And in the very first state, we have a description and in the second sheet we have this entire data.

30
00:02:40,110 --> 00:02:44,320
So here we have to basically parse the index of that sheet.

31
00:02:44,550 --> 00:02:52,410
So my index is basically one because zero indexes goals for description and the first one is goals for

32
00:02:52,440 --> 00:02:53,150
data as well.

33
00:02:53,790 --> 00:02:55,640
So it will basically return me.

34
00:02:56,250 --> 00:03:03,450
So I'm going to store it in detail and let's say I'm going to call ahead on my frame.

35
00:03:03,930 --> 00:03:12,270
So it will basically return me data, have some number of rules and some column and to check what are

36
00:03:12,270 --> 00:03:20,010
the rules and columns in your data, you will see somewhere around five thousand rows and columns.

37
00:03:20,020 --> 00:03:27,240
And to check whether I have any null values, indeed or not, you can call it is null, dark, some

38
00:03:27,600 --> 00:03:32,610
to get all your some of the any values in your data.

39
00:03:33,060 --> 00:03:41,550
So now you will see over here you don't have any of the missing values in your data, but you will see.

40
00:03:41,910 --> 00:03:43,500
Yeah, Aidy.

41
00:03:43,530 --> 00:03:48,960
And the zip code doesn't make sense for the analysis purposes.

42
00:03:49,110 --> 00:03:56,720
So I can simply remove these features or I can simply drop both of the features for this.

43
00:03:56,730 --> 00:03:58,520
I'm going to say the drop.

44
00:03:58,740 --> 00:04:06,060
And in this list I have to mention my board, the features, the very first one is my ID and the next

45
00:04:06,060 --> 00:04:09,410
one is basically my zip code.

46
00:04:09,660 --> 00:04:14,250
So you can either copy from here or you can set as manually.

47
00:04:14,580 --> 00:04:21,860
And now you have to set your X's parameter because I had to remove this as entire column.

48
00:04:22,200 --> 00:04:28,060
Then if you want to factor in this parameter, which is responsible for a bit in your data frame.

49
00:04:28,530 --> 00:04:32,370
So now you can set it as well to just execute it.

50
00:04:32,370 --> 00:04:33,600
If you are going to check.

51
00:04:33,930 --> 00:04:36,970
What are the different columns right now available in my data.

52
00:04:37,290 --> 00:04:40,160
So you will see I don't have any idea.

53
00:04:40,170 --> 00:04:44,530
And the zip code features right now available in my data.

54
00:04:44,820 --> 00:04:51,110
Let's say for Advanced ViSalus, I'm going to import my plotless library.

55
00:04:51,120 --> 00:04:59,520
So import floridly Dot Xpress to step and I'm going to create a elías as B.

56
00:05:00,500 --> 00:05:03,990
So now my property is that important in my cell.

57
00:05:04,010 --> 00:05:11,030
So let's open the assignment and this is exactly the assignment, Larry first was in my data preparation

58
00:05:11,330 --> 00:05:16,100
in which I have to go through with a very basic staff checking the missing values.

59
00:05:16,320 --> 00:05:21,570
I reviewed all the features that are you know, that doesn't make sense at all.

60
00:05:21,590 --> 00:05:29,900
And the second one is basically I have to use five points, some concept to get a description about

61
00:05:29,900 --> 00:05:31,110
your data.

62
00:05:31,520 --> 00:05:35,200
So what exactly is five point summary?

63
00:05:35,210 --> 00:05:42,620
The very first one is in five point summary is my minimum value, which stands for the zero percentile

64
00:05:42,750 --> 00:05:43,140
data.

65
00:05:43,340 --> 00:05:50,270
And the second one is basically my 20 percent later and the third one is my 50 percent on data, which

66
00:05:50,270 --> 00:05:51,960
is termed as median.

67
00:05:52,280 --> 00:05:57,410
The third one is basically my seventy five percent, which is termed as two three Esbjorn.

68
00:05:57,710 --> 00:06:05,600
And the last one is of course my maximum value, or you can say as one hundred percent on data so you

69
00:06:05,600 --> 00:06:12,260
can achieve all this concept by just calling box brought on your data.

70
00:06:12,500 --> 00:06:19,270
So for this, I'm going to use floridly in such scenarios for interactive results.

71
00:06:19,280 --> 00:06:27,410
So from plotless, you have to call a function known as Bonks, which gives me a beautiful box plot.

72
00:06:27,590 --> 00:06:33,110
So if you are going to pass a step, the very first one parameter is data frame and then you have to

73
00:06:33,110 --> 00:06:39,060
pass your data on X and Y axis depending upon what exactly you want.

74
00:06:39,440 --> 00:06:41,370
So the very first one is my data.

75
00:06:41,720 --> 00:06:49,580
So my data frame is basically the X and then on X axis I have to pass various data or you can pass it

76
00:06:49,580 --> 00:06:51,330
on Y axis as well.

77
00:06:51,710 --> 00:07:00,410
So let's say I am going to set my y axis data, so I'm going to use five point summary for, let's say

78
00:07:00,410 --> 00:07:07,370
all these features just compute from here and just let's say based on here.

79
00:07:07,820 --> 00:07:14,480
So these are exactly my let's say I'm going to remove this coverage, which is responsible for credit

80
00:07:14,480 --> 00:07:15,620
card average.

81
00:07:16,400 --> 00:07:21,640
And now I have five features over here and I'm going to visualize this.

82
00:07:21,860 --> 00:07:23,830
So let's say this is my finger.

83
00:07:23,930 --> 00:07:32,180
I'm going to store it in seven and I have to just call a show on this finger to visualize my beautiful

84
00:07:32,210 --> 00:07:34,270
or interactive box plot.

85
00:07:34,700 --> 00:07:37,490
So it will take a couple of seconds to execute.

86
00:07:37,520 --> 00:07:40,330
So this is exactly my beautiful box.

87
00:07:40,660 --> 00:07:47,000
So if you are going to hover your mouse, so now you will see for income, I have minimum value eight,

88
00:07:47,030 --> 00:07:54,410
which is this one and two one, which is my basically twenty five on data is basically thirty nine and

89
00:07:54,410 --> 00:08:00,690
median is sixty four, which is basically my 58 percent data and two three which is my seventy five

90
00:08:00,710 --> 00:08:05,140
percent and it stands for two three as ninety eight.

91
00:08:05,450 --> 00:08:11,590
And this is exactly my maximum value, which is my two to four.

92
00:08:11,780 --> 00:08:20,000
So these all are my five point summary for my data, similarly for a similarly for experience, similarly

93
00:08:20,000 --> 00:08:24,800
for family and similarly in case of education as well.

94
00:08:25,190 --> 00:08:32,120
So that's a brief idea how you can visualize your visual, how you can conclude your visas and what

95
00:08:32,120 --> 00:08:36,690
type of visas you guys can basically perform on your data.

96
00:08:36,710 --> 00:08:44,240
So if you have to conclude this visa, you can see this five point summary suggests that experience

97
00:08:44,240 --> 00:08:46,000
has some negative values.

98
00:08:46,280 --> 00:08:51,470
So we have to fix this because experience can't have negative values.

99
00:08:51,830 --> 00:08:58,570
So we can see the minimum max me and the standard deviation for all the other features as well.

100
00:08:58,910 --> 00:09:06,110
And income you will see income has too much noise and it is slightly skewed, right?

101
00:09:06,410 --> 00:09:11,390
It means it is high positive outliers in income.

102
00:09:11,750 --> 00:09:19,340
And if you are going to conclude this age and experience, so you will see age and experience are some

103
00:09:19,360 --> 00:09:21,860
already equally distributed.

104
00:09:21,920 --> 00:09:25,130
So that's a type of conclusion you can draw on.

105
00:09:25,130 --> 00:09:28,310
You can in sense, from this box plot.

106
00:09:28,320 --> 00:09:31,400
So that's all of our discussions in the upcoming session.

107
00:09:31,640 --> 00:09:37,850
We are going to deal with some of the analysis as well as this experience, because it contains right

108
00:09:37,850 --> 00:09:39,410
now some negative values.

109
00:09:39,800 --> 00:09:41,610
So hopefully you will love this session.

110
00:09:41,630 --> 00:09:42,280
Thank you.

111
00:09:42,290 --> 00:09:43,150
Have a nice day.

112
00:09:43,370 --> 00:09:46,700
Keep learning, keep going and keep motivating.