Would you like to inspect the original subtitles? These are the user uploaded subtitles that are being translated:
0
00:00:00,000 --> 00:00:01,042
PETER REDDIEN: All right.
1
00:00:01,042 --> 00:00:04,200
So let's turn now to some probability to go through this.
2
00:00:04,200 --> 00:00:07,510
3
00:00:07,510 --> 00:00:08,010
OK.
4
00:00:08,010 --> 00:00:10,540
5
00:00:10,540 --> 00:00:17,980
So this will be part three of the lecture called
6
00:00:17,980 --> 00:00:18,760
Bayes' theorem.
7
00:00:18,760 --> 00:00:26,740
8
00:00:26,740 --> 00:00:28,660
OK.
9
00:00:28,660 --> 00:00:31,900
So let's have some sample space here.
10
00:00:31,900 --> 00:00:40,190
11
00:00:40,190 --> 00:00:45,360
OK, which is going to be a collection of all
12
00:00:45,360 --> 00:00:49,050
possible discrete events or discrete outcomes.
13
00:00:49,050 --> 00:01:11,910
14
00:01:11,910 --> 00:01:12,430
OK.
15
00:01:12,430 --> 00:01:13,972
So that's our sample space collection
16
00:01:13,972 --> 00:01:16,000
of all possible discrete outcomes,
17
00:01:16,000 --> 00:01:22,780
and then we'll have some events, where
18
00:01:22,780 --> 00:01:25,600
an event is a collection of points in this sample space.
19
00:01:25,600 --> 00:01:44,960
20
00:01:44,960 --> 00:01:45,460
OK.
21
00:01:45,460 --> 00:01:47,293
So let's go through this with some examples.
22
00:01:47,293 --> 00:01:54,700
I'll give you an event, event A. OK.
23
00:01:54,700 --> 00:01:57,430
We'll imagine we're rolling dice.
24
00:01:57,430 --> 00:01:59,980
And we've rolled the dice.
25
00:01:59,980 --> 00:02:06,190
And we get a 1, a 2, a 4, or a 6.
26
00:02:06,190 --> 00:02:10,570
27
00:02:10,570 --> 00:02:13,550
A 1, 2, a 4, or a 6.
28
00:02:13,550 --> 00:02:14,050
OK.
29
00:02:14,050 --> 00:02:18,460
So we could put those outcomes possible outcomes in our sample
30
00:02:18,460 --> 00:02:23,020
space here, 1, 2, 4, and 6, and then
31
00:02:23,020 --> 00:02:26,470
I'll just circle those and say that's event
32
00:02:26,470 --> 00:02:32,760
A. That one of those discrete outcomes happens.
33
00:02:32,760 --> 00:02:35,088
So that'll be A.
34
00:02:35,088 --> 00:02:49,870
And now I'll give you an event B, where we get a 3 or a 6.
35
00:02:49,870 --> 00:02:56,620
36
00:02:56,620 --> 00:02:57,120
OK.
37
00:02:57,120 --> 00:03:00,390
So we get a 3, or a 6.
38
00:03:00,390 --> 00:03:06,930
So that's our event B. And we could also
39
00:03:06,930 --> 00:03:12,210
have obtained a 5 on the dice as an outcome,
40
00:03:12,210 --> 00:03:17,030
but that was neither event A or event B. OK.
41
00:03:17,030 --> 00:03:20,000
42
00:03:20,000 --> 00:03:22,400
So now we can ask some things about the probability
43
00:03:22,400 --> 00:03:24,330
of these events.
44
00:03:24,330 --> 00:03:26,850
So let's start easy here.
45
00:03:26,850 --> 00:03:30,350
So we just say the probability of event A happening.
46
00:03:30,350 --> 00:03:32,940
47
00:03:32,940 --> 00:03:35,100
So there's six possible outcomes.
48
00:03:35,100 --> 00:03:37,170
Event A has four of those outcomes.
49
00:03:37,170 --> 00:03:38,795
So the probability of event A is 4/6.
50
00:03:38,795 --> 00:03:42,470
51
00:03:42,470 --> 00:03:45,645
Probability of event B 2/6.
52
00:03:45,645 --> 00:03:52,030
53
00:03:52,030 --> 00:03:52,530
OK.
54
00:03:52,530 --> 00:03:54,900
So now we could say, what's the probability
55
00:03:54,900 --> 00:03:57,870
of event A or event B?
56
00:03:57,870 --> 00:04:05,760
57
00:04:05,760 --> 00:04:06,780
OK.
58
00:04:06,780 --> 00:04:09,570
Now you recall in the past we've talked a little bit
59
00:04:09,570 --> 00:04:15,090
about probabilities of two events happening, where
60
00:04:15,090 --> 00:04:16,470
it could be one or the other.
61
00:04:16,470 --> 00:04:19,779
If they are mutually exclusive independent events,
62
00:04:19,779 --> 00:04:21,779
then you just add the probabilities of those two
63
00:04:21,779 --> 00:04:23,400
things happening together.
64
00:04:23,400 --> 00:04:26,950
But here, we see that they are not mutually exclusive.
65
00:04:26,950 --> 00:04:32,400
So if we just added the probability of A
66
00:04:32,400 --> 00:04:35,130
to the probability of B, we would
67
00:04:35,130 --> 00:04:38,850
be double counting the probability of this outcome 6.
68
00:04:38,850 --> 00:04:46,168
So we now need to subtract the probability of A and B
69
00:04:46,168 --> 00:04:48,585
to account for this fact that we would be double counting.
70
00:04:48,585 --> 00:04:54,220
71
00:04:54,220 --> 00:05:08,360
So this is 4/6 plus 2/6 minus 1/6 equals 5/6.
72
00:05:08,360 --> 00:05:10,770
73
00:05:10,770 --> 00:05:12,770
And when you just look at it in the sample space
74
00:05:12,770 --> 00:05:16,520
is sort of pretty obvious that five out of these six outcomes
75
00:05:16,520 --> 00:05:19,830
would be A or B. OK.
76
00:05:19,830 --> 00:05:22,340
77
00:05:22,340 --> 00:05:26,960
Now we could calculate a conditional probability, where
78
00:05:26,960 --> 00:05:33,710
we say what is the probability of A given B.
79
00:05:33,710 --> 00:05:36,035
And I'll denote this with this vertical line here.
80
00:05:36,035 --> 00:05:40,220
81
00:05:40,220 --> 00:05:47,840
For this vertical line indicates given,
82
00:05:47,840 --> 00:05:53,780
where in other words, if we know that event B has happened,
83
00:05:53,780 --> 00:05:57,257
what is the probability that A happened?
84
00:05:57,257 --> 00:05:58,840
Can anyone tell me what that would be?
85
00:05:58,840 --> 00:06:01,330
Yeah.
86
00:06:01,330 --> 00:06:05,230
So if we event B happened, we know we either got a 3 or a 6,
87
00:06:05,230 --> 00:06:07,105
there's a 50% chance it was a 6.
88
00:06:07,105 --> 00:06:11,600
89
00:06:11,600 --> 00:06:12,100
OK.
90
00:06:12,100 --> 00:06:17,200
91
00:06:17,200 --> 00:06:22,120
Go to probability of event B given A. What would that be?
92
00:06:22,120 --> 00:06:25,650
93
00:06:25,650 --> 00:06:26,150
1/4.
94
00:06:26,150 --> 00:06:30,580
95
00:06:30,580 --> 00:06:34,500
If we know that A happened, you got a 1, 2, 4, 6.
96
00:06:34,500 --> 00:06:36,500
One of those four options would have been the 6.
97
00:06:36,500 --> 00:06:42,100
So 1/4 chance we got event B given we got event A. OK.
98
00:06:42,100 --> 00:06:51,083
99
00:06:51,083 --> 00:06:52,750
Now we could ask, what's the probability
100
00:06:52,750 --> 00:07:03,520
of event A and event B. And this will
101
00:07:03,520 --> 00:07:05,860
be equal to the probability of event
102
00:07:05,860 --> 00:07:16,000
A given B times the probability of event B.
103
00:07:16,000 --> 00:07:20,210
104
00:07:20,210 --> 00:07:26,240
So if B happened, then we had a 1/2 chance of getting A.
105
00:07:26,240 --> 00:07:28,032
But there is only a 2/6 chance of B
106
00:07:28,032 --> 00:07:29,240
happening in the first place.
107
00:07:29,240 --> 00:07:34,500
108
00:07:34,500 --> 00:07:38,900
So that's the probability of event A and B happening.
109
00:07:38,900 --> 00:07:43,730
You could also do this as the probability of event B given
110
00:07:43,730 --> 00:07:55,880
A times the probability of A. So probability B given a was 1/4,
111
00:07:55,880 --> 00:07:57,440
and the probability of A was 4/6.
112
00:07:57,440 --> 00:08:01,910
113
00:08:01,910 --> 00:08:03,860
OK.
114
00:08:03,860 --> 00:08:05,864
Equals 1/6.
115
00:08:05,864 --> 00:08:09,248
You could just see it by looking at our sample space there.
116
00:08:09,248 --> 00:08:11,540
You say, the probability of getting A and B is only one
117
00:08:11,540 --> 00:08:14,030
of those six events.
118
00:08:14,030 --> 00:08:15,740
So that's how it's sort of intuitive
119
00:08:15,740 --> 00:08:17,980
what I just went through.
7665
Can't find what you're looking for?
Get subtitles in any language from opensubtitles.com, and translate them here.