Would you like to inspect the original subtitles? These are the user uploaded subtitles that are being translated:
0
00:00:00,000 --> 00:00:02,760
PETER REDDIEN: What we're going to do
1
00:00:02,760 --> 00:00:07,770
is now try to set up a situation where
2
00:00:07,770 --> 00:00:09,960
we can get new information about probability,
3
00:00:09,960 --> 00:00:13,270
given new data or more information.
4
00:00:13,270 --> 00:00:16,740
So let's just consider some events here,
5
00:00:16,740 --> 00:00:27,190
back to our individual K. OK, so we
6
00:00:27,190 --> 00:00:33,640
could define some events here and say event
7
00:00:33,640 --> 00:00:47,260
A is that L is unaffected, and event B
8
00:00:47,260 --> 00:00:50,140
will be what we care about, that K is a carrier.
9
00:00:50,140 --> 00:01:02,270
10
00:01:02,270 --> 00:01:05,030
So what we really want to know is
11
00:01:05,030 --> 00:01:08,585
what is the probability of event B, given event A?
12
00:01:08,585 --> 00:01:12,200
That was the question I set up here.
13
00:01:12,200 --> 00:01:14,573
OK, so we want to derive an equation that
14
00:01:14,573 --> 00:01:17,240
would allow us to calculate that from these simple relationships
15
00:01:17,240 --> 00:01:20,360
we've gone through here.
16
00:01:20,360 --> 00:01:23,210
So what we're going to do is we're
17
00:01:23,210 --> 00:01:28,100
going to set this equation equal to that, because they're
18
00:01:28,100 --> 00:01:32,960
both equal to the probability of getting A and B.
19
00:01:32,960 --> 00:01:40,860
So we'll say that the probability
20
00:01:40,860 --> 00:01:47,520
of B given A times the probability of A is equal to--
21
00:01:47,520 --> 00:02:00,966
22
00:02:00,966 --> 00:02:04,891
the probability of B given A times the probability of A,
23
00:02:04,891 --> 00:02:07,270
that's this one, is equal to the probability
24
00:02:07,270 --> 00:02:09,460
of A given B times the probability of B.
25
00:02:09,460 --> 00:02:12,640
That's this one.
26
00:02:12,640 --> 00:02:14,640
So then we can say the probability of B
27
00:02:14,640 --> 00:02:19,950
given A is equal to the probability of A given
28
00:02:19,950 --> 00:02:26,550
B times the probability of B divided
29
00:02:26,550 --> 00:02:44,080
by the probability of A. This is Bayes' theorem, which
30
00:02:44,080 --> 00:02:49,130
is one of the most widely used relations in statistics today.
31
00:02:49,130 --> 00:03:12,140
This will be the posterior probability,
32
00:03:12,140 --> 00:03:16,025
the new probability given our data, our information A.
33
00:03:16,025 --> 00:03:24,320
And this is the prior probability, the probability
34
00:03:24,320 --> 00:03:28,550
of event B before we knew this information about A.
35
00:03:28,550 --> 00:03:31,070
So using this, we can now get data
36
00:03:31,070 --> 00:03:32,630
and update our probability of event
37
00:03:32,630 --> 00:03:35,780
B. It's called conditional probability.
38
00:03:35,780 --> 00:03:38,300
39
00:03:38,300 --> 00:03:44,360
So now let's apply it to our question about scenario
40
00:03:44,360 --> 00:03:52,010
K. Or, sorry, event A and B with K being a carrier or not.
41
00:03:52,010 --> 00:03:55,910
42
00:03:55,910 --> 00:04:05,480
All right, so our probability of event B, the K is a carrier,
43
00:04:05,480 --> 00:04:08,048
was 0.5.
44
00:04:08,048 --> 00:04:09,215
We went through that before.
45
00:04:09,215 --> 00:04:12,590
46
00:04:12,590 --> 00:04:23,840
Now our probability of A given B is one half.
47
00:04:23,840 --> 00:04:28,580
48
00:04:28,580 --> 00:04:33,060
If K is a carrier, there's a one half chance of transmission.
49
00:04:33,060 --> 00:04:41,600
Now what's our probability of A, the probability
50
00:04:41,600 --> 00:04:46,550
that L is unaffected, prior to having any data.
51
00:04:46,550 --> 00:04:48,590
Well, it could be that K was a carrier,
52
00:04:48,590 --> 00:04:51,470
or it could be the K was not a carrier.
53
00:04:51,470 --> 00:04:54,260
So the probability of A will be the probability
54
00:04:54,260 --> 00:04:58,370
of A given B times the probability
55
00:04:58,370 --> 00:05:06,320
of A times the probability of B plus the probability
56
00:05:06,320 --> 00:05:15,740
of A given not B times the probability of not B.
57
00:05:15,740 --> 00:05:19,340
What we're doing here is just considering the two scenarios
58
00:05:19,340 --> 00:05:21,440
that K is a carrier, or K is not a carrier.
59
00:05:21,440 --> 00:05:25,920
In both cases, what would the probability of getting event A
60
00:05:25,920 --> 00:05:26,420
be?
61
00:05:26,420 --> 00:05:29,660
62
00:05:29,660 --> 00:05:30,605
So this is 3/4.
63
00:05:30,605 --> 00:05:35,730
64
00:05:35,730 --> 00:05:39,920
So now we can calculate our probability
65
00:05:39,920 --> 00:05:40,775
that K is a carrier.
66
00:05:40,775 --> 00:05:53,060
67
00:05:53,060 --> 00:06:00,260
One half times one half divided by 3/4 equals one third.
68
00:06:00,260 --> 00:06:05,800
69
00:06:05,800 --> 00:06:09,430
So given the information that L would be unaffected,
70
00:06:09,430 --> 00:06:12,190
now the probability that K is a carrier is no longer one half,
71
00:06:12,190 --> 00:06:13,090
it's one third.
72
00:06:13,090 --> 00:06:14,240
It has dropped.
73
00:06:14,240 --> 00:06:17,590
And you could calculate that for any type of event
74
00:06:17,590 --> 00:06:19,530
using this relation.
5318
Can't find what you're looking for?
Get subtitles in any language from opensubtitles.com, and translate them here.