1
00:00:00,330 --> 00:00:06,000
Hello and welcome back to another class of our course about the complete introduction to science, as

2
00:00:06,000 --> 00:00:12,480
well as that analysis with the use of Python in this class, we are going to talk about the really important

3
00:00:12,480 --> 00:00:18,780
part of the science, which would be statistics or basically statistics are not the base of data science,

4
00:00:18,780 --> 00:00:22,010
but are a huge player in data science.

5
00:00:22,500 --> 00:00:29,070
Indeed, a lot of techniques, a lot of statistical measures are used in data science, and this is

6
00:00:29,070 --> 00:00:30,220
why it's really important.

7
00:00:30,540 --> 00:00:33,930
So in this class, we'll just have a small introduction to statistics.

8
00:00:33,930 --> 00:00:40,590
We're not going to talk about all the statistics, terminologies and all the statistics measures that

9
00:00:40,590 --> 00:00:43,430
are used, but just have an introduction to statistics.

10
00:00:43,460 --> 00:00:49,170
What is it exactly where it can be used in the future classes?

11
00:00:49,170 --> 00:00:55,860
We are going to talk about the basics that you guys would need to have in statistics to be able to understand

12
00:00:55,860 --> 00:00:56,450
that science.

13
00:00:56,820 --> 00:00:57,750
So let's start.

14
00:00:58,710 --> 00:01:05,730
So basically what the statistics say, if I go with a basic explanation of what is statistics, that

15
00:01:05,740 --> 00:01:12,480
simply a branch of maths that will deal with the collection, analysis, interpretation and the representation

16
00:01:12,900 --> 00:01:15,570
of masses of numerical data.

17
00:01:15,810 --> 00:01:22,710
So basically, you take a population and you will work with, let's say, a sample of this population.

18
00:01:22,950 --> 00:01:25,470
Indeed, it's possible to work with the whole population.

19
00:01:25,470 --> 00:01:32,550
But once again, this could be really complicated because it's not always possible to, let's say,

20
00:01:32,550 --> 00:01:35,690
for example, make service to the whole population.

21
00:01:35,970 --> 00:01:38,950
So we will work with samples of population.

22
00:01:38,970 --> 00:01:43,410
So let's say, for example, we want to work with, I don't know, with all the American population

23
00:01:44,040 --> 00:01:47,340
to know what is their favorite color, for example.

24
00:01:48,030 --> 00:01:53,610
So instead of asking all the papillae, all the American population will take different sample.

25
00:01:53,630 --> 00:01:58,220
So we're not going to talk about sampling in this class, but this is just an example.

26
00:01:58,410 --> 00:02:02,380
So you can take a sample, for example, from each city in each state.

27
00:02:02,400 --> 00:02:04,190
So this is just one example.

28
00:02:04,500 --> 00:02:10,140
And from that sample, since it's random people, you guys will be able we will be able to bring to

29
00:02:10,140 --> 00:02:15,750
come to a conclusion that the favorite color of this sample is red, to say that the favorite color

30
00:02:15,750 --> 00:02:19,920
of American based on a certain sample is red.

31
00:02:20,790 --> 00:02:24,720
So this is, for example, this is one example of how statistics could be used.

32
00:02:24,730 --> 00:02:31,680
And as you can see, this is some really basic data science things that we have here, statistics to

33
00:02:31,690 --> 00:02:32,940
be used in many fields.

34
00:02:32,970 --> 00:02:35,920
So as you can see here, it could be used in education.

35
00:02:36,000 --> 00:02:42,410
Know, for example, I don't know which well, we can use statistics to see, for example, which in

36
00:02:42,420 --> 00:02:49,710
which type of classes students perform the most or perform the less and try to find out the problems

37
00:02:49,710 --> 00:02:51,930
in, for example, those classes.

38
00:02:51,940 --> 00:02:55,870
For example, let's say students have problems in maths in this field.

39
00:02:55,890 --> 00:02:56,490
Exactly.

40
00:02:56,490 --> 00:03:00,100
Then we will try to find out the problem, why students have problems in this field.

41
00:03:00,120 --> 00:03:03,450
Maybe it's because of the way it's presented.

42
00:03:03,450 --> 00:03:09,070
Maybe it's because, I don't know, students are just not motivated by this field.

43
00:03:09,100 --> 00:03:10,710
Maybe there is any.

44
00:03:10,740 --> 00:03:13,500
Well, we'll try to find the explanation to this.

45
00:03:13,770 --> 00:03:18,060
I mean, it's possible to do it with statistics as well as data science.

46
00:03:18,060 --> 00:03:23,670
But once again, statistics will be the basics of all this could be used in stock market life science,

47
00:03:23,670 --> 00:03:27,710
whether retail insurances and many, many other fields.

48
00:03:27,930 --> 00:03:29,430
Once again, it's not used alone.

49
00:03:29,430 --> 00:03:33,540
It's used with other mathematical fields as well as data science.

50
00:03:35,220 --> 00:03:38,460
So right now, we'll talk about the types of statistics.

51
00:03:38,460 --> 00:03:44,160
Basically, we have two types of statistics, descriptive statistics and inferential statistics.

52
00:03:44,820 --> 00:03:48,000
We are not going to come into details of each of them.

53
00:03:48,000 --> 00:03:54,810
But I'll try to give you a brief definition of what are those two types of statistics and just give

54
00:03:54,810 --> 00:03:56,740
you a small example of what it could be.

55
00:03:57,630 --> 00:04:04,260
So basically, for the descriptive statistic, it will help us organize the data and focus on the character

56
00:04:04,260 --> 00:04:06,120
of data providing parameters.

57
00:04:07,140 --> 00:04:14,610
So this would be for the descriptive statistic, and it's really used to well, it's the use of data

58
00:04:14,610 --> 00:04:22,420
to provide a description of people of a population sort either through numerical or graphs or tables.

59
00:04:22,770 --> 00:04:27,360
So this is basically for descriptive statistics and inferential statistics.

60
00:04:27,390 --> 00:04:28,710
It's a bit different.

61
00:04:29,130 --> 00:04:37,620
It's simply generalized generalized data set and applying probabilities to arrive to a conclusion.

62
00:04:38,520 --> 00:04:44,980
And the inferential statistics will make inferences and predictions about population based on a same

63
00:04:45,210 --> 00:04:48,450
sample of data taken from a certain population.

64
00:04:49,140 --> 00:04:53,060
So basically, let's say that we have wine.

65
00:04:53,070 --> 00:04:58,770
So let's say you guys are producing wine in huge quantities, quantities, and you guys want to know

66
00:04:58,780 --> 00:05:01,980
what you guys want to look at the age of your wine.

67
00:05:02,250 --> 00:05:07,950
So basically with the descriptive statistics that when you guys will start the age of your wine, you

68
00:05:07,950 --> 00:05:09,320
will have the average.

69
00:05:09,330 --> 00:05:11,460
So you will have the average age of your wine.

70
00:05:12,210 --> 00:05:13,480
You will have the max.

71
00:05:13,500 --> 00:05:22,200
So the max, the max, the highest age of your wine, the min, so the lowest and the wine that is the

72
00:05:22,200 --> 00:05:23,090
less aged.

73
00:05:24,120 --> 00:05:29,790
And basically divide it into three categories like this, inferential statistics.

74
00:05:29,800 --> 00:05:34,140
It's going to be a bit well, it's going to be a bit different.

75
00:05:34,560 --> 00:05:39,930
You will still have your average in the center, but you will have a category that will be, in this

76
00:05:39,930 --> 00:05:44,040
case, aged wines, not max aged wine.

77
00:05:44,040 --> 00:05:50,220
So you will just have aged wine where you will have all the winds that are aged more than a certain,

78
00:05:50,280 --> 00:05:53,850
let's say, age, and you will have less aged wines.

79
00:05:54,070 --> 00:05:55,770
So in this case, you will have, let's say.

80
00:05:57,090 --> 00:06:04,350
All the ones that are less age than a certain age in this case will work more with the quartiles instead

81
00:06:04,350 --> 00:06:06,870
of just working with Maxims and Minimum's.

82
00:06:07,380 --> 00:06:09,600
You can also apply this to, say, population.

83
00:06:09,600 --> 00:06:13,560
So a population that is short or small while a population.

84
00:06:13,570 --> 00:06:15,800
So you want to know the height of a certain population.

85
00:06:16,080 --> 00:06:21,950
So in the descriptive statistics, you will have the highest person, you will have the max height,

86
00:06:22,260 --> 00:06:26,610
so you will have the highest person in the room and the shortest person in the room and you will have

87
00:06:26,610 --> 00:06:27,450
the average.

88
00:06:27,720 --> 00:06:32,880
So basically you have those three persons and inferential statistics.

89
00:06:32,880 --> 00:06:38,400
You will have a category that will be tall, a category that would be short, and you have the average

90
00:06:38,400 --> 00:06:39,210
as well.

91
00:06:40,170 --> 00:06:43,200
So basically, those are for the two types of statistics.

92
00:06:43,890 --> 00:06:47,820
I'm then why statistics are useful in data science.

93
00:06:47,820 --> 00:06:51,300
So basically there are really useful for plenty of reasons.

94
00:06:51,300 --> 00:06:58,440
But for me personally, I think that the two main reasons why they are really useful is that it will

95
00:06:58,440 --> 00:07:04,710
give you a basic understanding of different ablazing methods and techniques as well.

96
00:07:05,070 --> 00:07:12,150
So basically, data well, statistics brings to data science many, many techniques and methods that

97
00:07:12,150 --> 00:07:16,710
could be used in data science to be able to, well, analyze.

98
00:07:16,860 --> 00:07:18,480
We better the data.

99
00:07:19,350 --> 00:07:23,900
So basically, this is why your statistics are really, really important data science.

100
00:07:24,090 --> 00:07:28,290
And as I said, statistics are really the base of data science.

101
00:07:28,680 --> 00:07:32,230
And this is well, this is why it's really important.

102
00:07:32,550 --> 00:07:39,900
And the second reason is because it allows data scientists to use many measures such as?

103
00:07:40,020 --> 00:07:41,840
Well, those are just some of them.

104
00:07:42,030 --> 00:07:47,790
So we'll have the merrian the mean, the standard deviation and, well, many more to be able to make

105
00:07:47,790 --> 00:07:50,580
a complete analysis of something.

106
00:07:50,620 --> 00:07:54,530
Well, when I see something good, it could be, for example, in the financial field.

107
00:07:54,840 --> 00:07:59,100
So let's say you want to analyze a stock, you can use the standard deviation.

108
00:07:59,100 --> 00:08:03,150
And this is something that is really, really important because, well, for example, for a stock,

109
00:08:03,480 --> 00:08:06,480
a high standard deviation means that the stock is very volatile.

110
00:08:06,480 --> 00:08:10,110
It's a low standard deviation means that the stock will not move that much.

111
00:08:10,470 --> 00:08:17,250
And once again, this is one this is something that comes from statistics and that can be used in data

112
00:08:17,250 --> 00:08:17,730
science.

113
00:08:17,730 --> 00:08:22,370
Once again, the median, for example, I want to find out the median of something I want to find on

114
00:08:22,410 --> 00:08:24,830
the mean or the average of something.

115
00:08:25,590 --> 00:08:27,050
Those are all good.

116
00:08:27,060 --> 00:08:33,470
That's all coming from statistics and is very, very used into data science.

117
00:08:33,900 --> 00:08:39,990
So what you see here are just some graphs of some analysis that have been made a while.

118
00:08:39,990 --> 00:08:42,960
You will be able to perform those types of analysis.

119
00:08:43,260 --> 00:08:48,300
Those are simply results of of analysis that have been made.

120
00:08:48,300 --> 00:08:52,740
I don't know what those analysis, but those are just the results of the results.

121
00:08:52,740 --> 00:08:53,790
Look, something like this.

122
00:08:54,210 --> 00:09:00,120
And at the end of the day, this is a mix of statistics and the science put it together to be able to

123
00:09:00,120 --> 00:09:07,860
analyze a population or a certain dataset, to be able to answer a certain problematic in this world

124
00:09:08,130 --> 00:09:12,120
and to come to a conclusion about, well, about something.

125
00:09:12,960 --> 00:09:17,370
And to solve a problem, so because this is the main goal of their science, as well as the statistics

126
00:09:17,370 --> 00:09:20,140
in their science to solve a certain problem.

127
00:09:20,640 --> 00:09:23,620
So you guys right now understand why statistics are that important.

128
00:09:24,150 --> 00:09:32,300
Now it's time for us to start learning all the basics that will be really important for data science,

129
00:09:32,310 --> 00:09:38,400
while all the statistical bases that will be really important in data science, for discourse and for

130
00:09:38,400 --> 00:09:41,180
your future, as it are saying this.

131
00:09:41,940 --> 00:09:42,800
So that's it for.

132
00:09:42,980 --> 00:09:43,440
Guys, guys.

133
00:09:43,440 --> 00:09:48,090
And see all in our next class where we are going to learn all those basics.