Would you like to inspect the original subtitles? These are the user uploaded subtitles that are being translated:
0
00:00:00,000 --> 00:00:01,958
PETER REDDIEN: All right.
1
00:00:01,958 --> 00:00:04,500
Now let's think about when-- if we look at these classes what
2
00:00:04,500 --> 00:00:07,678
we expect with our two hypotheses
3
00:00:07,678 --> 00:00:09,720
that we're considering, hypotheses two and three.
4
00:00:09,720 --> 00:00:23,610
5
00:00:23,610 --> 00:00:28,563
What's expected if we have these three classes--
6
00:00:28,563 --> 00:00:29,730
with these three hypotheses?
7
00:00:29,730 --> 00:00:31,470
We went through hypothesis one before.
8
00:00:31,470 --> 00:00:39,880
We're expecting a three to one ratio
9
00:00:39,880 --> 00:00:48,802
of not paralyzed to paralyzed.
10
00:00:48,802 --> 00:00:53,320
11
00:00:53,320 --> 00:00:55,910
So that's what we did in the first part of the lecture.
12
00:00:55,910 --> 00:01:02,290
So now hypothesis two, what do we expect?
13
00:01:02,290 --> 00:01:04,330
So recall hypothesis two.
14
00:01:04,330 --> 00:01:06,880
We need to be homozygous for both
15
00:01:06,880 --> 00:01:11,560
of these ts alleles of paralyzed gene and shibere gene.
16
00:01:11,560 --> 00:01:13,900
So what ratio of phenotypes do we expect here?
17
00:01:13,900 --> 00:01:17,460
18
00:01:17,460 --> 00:01:22,110
What's the ratio of not paralyzed to paralyzed?
19
00:01:22,110 --> 00:01:22,800
Yeah?
20
00:01:22,800 --> 00:01:23,760
STUDENT: 15 to one.
21
00:01:23,760 --> 00:01:27,120
PETER REDDIEN: 15 to one, OK.
22
00:01:27,120 --> 00:01:30,120
23
00:01:30,120 --> 00:01:33,210
We expect 15 to one because only this class would be paralyzed
24
00:01:33,210 --> 00:01:35,520
and it's 1/16.
25
00:01:35,520 --> 00:01:37,620
Finally, what would we expect in hypothesis three?
26
00:01:37,620 --> 00:01:43,390
27
00:01:43,390 --> 00:01:45,412
Yeah?
28
00:01:45,412 --> 00:01:46,370
STUDENT: Seven to nine.
29
00:01:46,370 --> 00:01:50,100
PETER REDDIEN: Yeah, nine to seven.
30
00:01:50,100 --> 00:01:50,600
Oops.
31
00:01:50,600 --> 00:01:53,810
32
00:01:53,810 --> 00:01:55,970
Nine not paralyzed to seven paralyzed.
33
00:01:55,970 --> 00:01:59,510
OK, great.
34
00:01:59,510 --> 00:02:02,480
Now we can look at our data and think
35
00:02:02,480 --> 00:02:05,930
about these different hypotheses.
36
00:02:05,930 --> 00:02:07,715
Our data, if you recall--
37
00:02:07,715 --> 00:02:11,490
38
00:02:11,490 --> 00:02:19,230
so the data was 12 to four.
39
00:02:19,230 --> 00:02:22,520
40
00:02:22,520 --> 00:02:24,380
Or three to one.
41
00:02:24,380 --> 00:02:28,610
But those are our actual numbers, 12 to four.
42
00:02:28,610 --> 00:02:31,648
What can we conclude?
43
00:02:31,648 --> 00:02:33,460
What do you guys think?
44
00:02:33,460 --> 00:02:33,960
Yeah?
45
00:02:33,960 --> 00:02:36,670
STUDENT: Hypothesis one is most likely.
46
00:02:36,670 --> 00:02:39,580
PETER REDDIEN: OK, you're saying hypothesis one is most likely?
47
00:02:39,580 --> 00:02:40,250
OK.
48
00:02:40,250 --> 00:02:41,710
Certainly the closest to the data.
49
00:02:41,710 --> 00:02:45,120
Can we rule these out?
50
00:02:45,120 --> 00:02:45,690
OK.
51
00:02:45,690 --> 00:02:48,490
Well, maybe, maybe not.
52
00:02:48,490 --> 00:02:51,516
How do we know if we can rule them out?
53
00:02:51,516 --> 00:02:53,125
What do you think?
54
00:02:53,125 --> 00:02:55,250
I mean, so-- well, 15 to one looks pretty different
55
00:02:55,250 --> 00:02:56,060
from 12 to four.
56
00:02:56,060 --> 00:03:00,140
But nine to seven, is that that different from 12 to four?
57
00:03:00,140 --> 00:03:01,640
I don't know, maybe you would expect
58
00:03:01,640 --> 00:03:03,560
to get that, some probability.
59
00:03:03,560 --> 00:03:06,310
60
00:03:06,310 --> 00:03:08,070
What do we do?
61
00:03:08,070 --> 00:03:10,590
We just live with more likely, or less likely, or-- yeah?
62
00:03:10,590 --> 00:03:12,057
STUDENT: A statistical test.
63
00:03:12,057 --> 00:03:14,033
PETER REDDIEN: Statistical test, great.
64
00:03:14,033 --> 00:03:15,950
So what we're going to do is a hypothesis test
65
00:03:15,950 --> 00:03:17,060
with a statistical test.
66
00:03:17,060 --> 00:03:18,477
And what we're going to ask, let's
67
00:03:18,477 --> 00:03:22,280
take hypothesis three, because it's the closest,
68
00:03:22,280 --> 00:03:28,610
and say, so with hypothesis three,
69
00:03:28,610 --> 00:03:32,630
how likely would it be to get data--
70
00:03:32,630 --> 00:03:35,960
if hypothesis three is correct, how likely
71
00:03:35,960 --> 00:03:40,640
would it be to get data that deviates this far or greater
72
00:03:40,640 --> 00:03:43,170
from the expected?
73
00:03:43,170 --> 00:03:45,350
That's what we're going to ask.
4700
Can't find what you're looking for?
Get subtitles in any language from opensubtitles.com, and translate them here.