|
1 |
|
00:00:06,760 |
|
Eventually I will give some practice problem for |
|
|
|
2 |
|
00:00:10,320 |
|
chapter eight. Generally speaking, there are three |
|
|
|
3 |
|
00:00:15,040 |
|
types of questions. The first type, multiple |
|
|
|
4 |
|
00:00:19,800 |
|
choice, so MCQ questions. |
|
|
|
5 |
|
00:00:36,250 |
|
The other type of problems will be true or false. |
|
|
|
6 |
|
00:00:42,890 |
|
Part B, Part C, three response problems. |
|
|
|
7 |
|
00:00:56,510 |
|
So three types of questions. Multiple choice, we |
|
|
|
8 |
|
00:01:00,210 |
|
have four answers. You have to select correct one. |
|
|
|
9 |
|
00:01:06,060 |
|
True or false problems. And the last part, free |
|
|
|
10 |
|
00:01:11,660 |
|
response problems. Here we'll talk about one of |
|
|
|
11 |
|
00:01:14,800 --> 00:01:20,740 |
|
these. I will cover multiple choice questions as |
|
|
|
12 |
|
00:01:20,740 --> 00:01:24,200 |
|
well as true and false. Let's start with number |
|
|
|
13 |
|
00:01:24,200 |
|
one for multiple choice. The width of a confidence |
|
|
|
14 |
|
00:01:30,200 |
|
interval estimate for a proportion will be Here we |
|
|
|
15 |
|
00:01:36,050 |
|
are talking about the width of a confidence |
|
|
|
16 |
|
00:01:39,690 |
|
interval. |
|
|
|
17 |
|
00:01:43,070 |
|
Estimates for a proportion will be narrower for 99 |
|
|
|
18 |
|
00:01:48,770 |
|
% confidence than for a 9%. For 95 confidence? No, |
|
|
|
19 |
|
00:01:56,280 |
|
because as we know that as the confidence level |
|
|
|
20 |
|
00:01:59,120 |
|
increases, the width becomes wider. So A is |
|
|
|
21 |
|
00:02:03,120 |
|
incorrect. Is this true? B. Wider for sample size |
|
|
|
22 |
|
00:02:10,440 |
|
of 100 than for a sample size of 50? False, |
|
|
|
23 |
|
00:02:15,020 |
|
because as sample size increases, The sampling |
|
|
|
24 |
|
00:02:20,400 |
|
error goes down. That means the width of the |
|
|
|
25 |
|
00:02:23,600 |
|
interval becomes smaller and smaller. Yes, for N. |
|
|
|
26 |
|
00:02:30,300 |
|
Part C. Normal for 90% confidence, then for 95% |
|
|
|
27 |
|
00:02:37,100 |
|
confidence. That's correct. So C is the correct |
|
|
|
28 |
|
00:02:40,620 --> 00:02:43,640 |
|
answer. Part D. Normal when the sampling |
|
|
|
29 |
|
00:02:43,640 --> 00:02:49,100 |
|
proportion is 50%. is incorrect because if we have |
|
|
|
30 |
|
00:02:49,100 --> 00:02:52,920 |
|
smaller than 50%, we'll get smaller confidence, |
|
|
|
31 |
|
00:02:53,320 |
|
smaller weight of the confidence. So C is the |
|
|
|
32 |
|
00:02:56,620 |
|
correct answer. Any question? So C is the correct |
|
|
|
33 |
|
00:03:02,720 |
|
answer because as C level increases, the |
|
|
|
34 |
|
00:03:06,180 |
|
confidence interval becomes narrower. |
|
|
|
35 |
|
00:03:11,040 |
|
Let's move to the second one. |
|
|
|
36 |
|
00:03:16,540 --> 00:03:19,900 |
|
A 99% confidence interval estimate can be |
|
|
|
37 |
|
00:03:19,900 --> 00:03:23,940 |
|
interpreted to mean that. Let's look at the |
|
|
|
38 |
|
00:03:23,940 |
|
interpretation of the 99% confidence interval. |
|
|
|
39 |
|
00:03:28,940 |
|
Part eight. |
|
|
|
40 |
|
00:03:33,360 |
|
If all possible samples are taken and confidence |
|
|
|
41 |
|
00:03:38,820 |
|
interval estimates are developed, 99% of them |
|
|
|
42 |
|
00:03:43,070 |
|
would include the true population mean somewhere |
|
|
|
43 |
|
00:03:45,750 |
|
within their interval. |
|
|
|
44 |
|
00:03:49,750 |
|
Here we are talking about the population mean. It |
|
|
|
45 |
|
00:03:53,210 |
|
says that 99% of them of these intervals would |
|
|
|
46 |
|
00:03:57,890 |
|
include the true population mean somewhere within |
|
|
|
47 |
|
00:04:00,790 |
|
their interval. It's correct. Why false? Why is it |
|
|
|
48 |
|
00:04:05,490 --> 00:04:10,620 |
|
false? This is correct answer, because it's |
|
|
|
49 |
|
00:04:10,620 |
|
mentioned that 99% of these confidence intervals |
|
|
|
50 |
|
00:04:15,240 |
|
will contain the true population mean somewhere |
|
|
|
51 |
|
00:04:19,600 |
|
within their interval. So A is correct. Let's look |
|
|
|
52 |
|
00:04:22,900 --> 00:04:30,880 |
|
at B. B says we have 99% confidence that we have |
|
|
|
53 |
|
00:04:30,880 --> 00:04:34,160 |
|
selected a sample whose interval does include the |
|
|
|
54 |
|
00:04:34,160 --> 00:04:39,160 |
|
population mean. Also, this one is correct. Again, |
|
|
|
55 |
|
00:04:39,300 --> 00:04:42,540 |
|
it's mentioned that 99% confidence that we have |
|
|
|
56 |
|
00:04:42,540 |
|
selected sample whose interval does include. So |
|
|
|
57 |
|
00:04:47,080 |
|
it's correct. So C is both of the above and D none |
|
|
|
58 |
|
00:04:52,600 --> 00:04:54,840 |
|
of the above. So C is the correct answer. So |
|
|
|
59 |
|
00:04:54,840 --> 00:04:59,080 |
|
sometimes maybe there is only one answer. Maybe in |
|
|
|
60 |
|
00:04:59,080 --> 00:05:03,360 |
|
other problems, it might be two answers are |
|
|
|
61 |
|
00:05:03,360 --> 00:05:07,150 |
|
correct. So for this one, B and C. I'm sorry, A |
|
|
|
62 |
|
00:05:07,150 |
|
and B are correct, so C is the correct answer. |
|
|
|
63 |
|
00:05:14,270 |
|
Number three, which of the following is not true |
|
|
|
64 |
|
00:05:17,530 |
|
about the student's T distribution? Here, we are |
|
|
|
65 |
|
00:05:20,610 --> 00:05:25,550 |
|
talking about the not true statement about the |
|
|
|
66 |
|
00:05:25,550 --> 00:05:31,110 |
|
student T distribution, A. It has more data in the |
|
|
|
67 |
|
00:05:31,110 --> 00:05:35,780 |
|
tails. and less in the center than does the normal |
|
|
|
68 |
|
00:05:35,780 --> 00:05:38,580 |
|
distribution. That's correct because we mentioned |
|
|
|
69 |
|
00:05:38,580 |
|
last time that T distribution, the tail is fatter |
|
|
|
70 |
|
00:05:44,160 |
|
than the Z normal. So that means it has more data |
|
|
|
71 |
|
00:05:48,000 |
|
in the tails and less data in the center. So |
|
|
|
72 |
|
00:05:52,100 |
|
that's correct. |
|
|
|
73 |
|
00:05:58,000 --> 00:06:01,020 |
|
It is used to construct confidence intervals for |
|
|
|
74 |
|
00:06:01,020 --> 00:06:03,220 |
|
the population mean when the population standard |
|
|
|
75 |
|
00:06:03,220 --> 00:06:07,400 |
|
deviation is known. No, we use z instead of t, so |
|
|
|
76 |
|
00:06:07,400 --> 00:06:11,680 |
|
this one is incorrect about t. It is well-shaped |
|
|
|
77 |
|
00:06:11,680 --> 00:06:17,320 |
|
and symmetrical, so that's true, so we are looking |
|
|
|
78 |
|
00:06:17,320 |
|
for the incorrect statement. D, as the number of |
|
|
|
79 |
|
00:06:21,900 |
|
degrees of freedom increases, |
|
|
|
80 |
|
00:06:25,850 |
|
The T distribution approaches the normal. That's |
|
|
|
81 |
|
00:06:31,510 --> 00:06:36,830 |
|
true. So which one? P. So P is incorrect. So |
|
|
|
82 |
|
00:06:36,830 --> 00:06:39,670 |
|
number four. Extra. |
|
|
|
83 |
|
00:06:42,010 --> 00:06:47,090 |
|
Can you explain the average total compensation of |
|
|
|
84 |
|
00:06:47,090 --> 00:06:51,830 |
|
CEOs in the service industry? Data were randomly |
|
|
|
85 |
|
00:06:51,830 --> 00:06:57,480 |
|
collected from 18 CEOs and 19 employees. 97% |
|
|
|
86 |
|
00:06:57,480 --> 00:07:06,040 |
|
confidence interval was calculated to be $281, |
|
|
|
87 |
|
00:07:07,040 --> 00:07:09,020 |
|
$260, |
|
|
|
88 |
|
00:07:10,060 --> 00:07:13,780 |
|
$5836, |
|
|
|
89 |
|
00:07:14,800 --> 00:07:19,300 |
|
and $180. Which of the following interpretations |
|
|
|
90 |
|
00:07:19,300 --> 00:07:27,310 |
|
is correct? Part number A. It says 97% of the |
|
|
|
91 |
|
00:07:27,310 --> 00:07:32,450 |
|
sample data compensation value between these two |
|
|
|
92 |
|
00:07:32,450 --> 00:07:37,310 |
|
values, correct or incorrect statement. Because it |
|
|
|
93 |
|
00:07:37,310 --> 00:07:44,310 |
|
says 97% of the sample data. For the confidence |
|
|
|
94 |
|
00:07:44,310 --> 00:07:48,310 |
|
center value, we are looking for the average, not |
|
|
|
95 |
|
00:07:48,310 --> 00:07:51,690 |
|
for the population, not for the sample. So A is |
|
|
|
96 |
|
00:07:51,690 --> 00:07:55,890 |
|
incorrect. Because A, it says here 97% of the |
|
|
|
97 |
|
00:07:55,890 --> 00:07:58,690 |
|
sampling total. Sample total, we are looking for |
|
|
|
98 |
|
00:07:58,690 --> 00:08:02,390 |
|
the average of the population. So A is incorrect |
|
|
|
99 |
|
00:08:02,390 --> 00:08:09,150 |
|
statement. B, we are 97% confident that the mean |
|
|
|
100 |
|
00:08:09,150 --> 00:08:15,890 |
|
of the sample. So it's false. Because the |
|
|
|
101 |
|
00:08:15,890 |
|
confidence about the entire population is about |
|
|
|
102 |
|
00:08:18,470 |
|
the population mean. So B is incorrect. C. In the |
|
|
|
103 |
|
00:08:24,160 |
|
population of the surface industry, here we have |
|
|
|
104 |
|
00:08:27,760 |
|
97% of them will have a total death toll. Also, |
|
|
|
105 |
|
00:08:33,360 |
|
this one is incorrect because it mentions in the |
|
|
|
106 |
|
00:08:37,480 |
|
population. Here we are talking about total, but |
|
|
|
107 |
|
00:08:39,900 |
|
we are looking for the average. Now, part D. We |
|
|
|
108 |
|
00:08:44,000 |
|
are 97% confident that the average total |
|
|
|
109 |
|
00:08:50,460 |
|
So this one is correct statement. So D is the |
|
|
|
110 |
|
00:08:53,440 |
|
correct statement. So for the confidence interval, |
|
|
|
111 |
|
00:08:55,840 |
|
we are looking for population, number one. Number |
|
|
|
112 |
|
00:08:59,040 |
|
two, the average of that population. So D is the |
|
|
|
113 |
|
00:09:03,520 |
|
correct answer. Let's go back to part A. In part |
|
|
|
114 |
|
00:09:07,260 --> 00:09:10,420 |
|
A, it says sample total. So this is incorrect. |
|
|
|
115 |
|
00:09:11,380 --> 00:09:15,140 |
|
Next one. The mean of the sample. We are looking |
|
|
|
116 |
|
00:09:15,140 --> 00:09:17,440 |
|
for the mean of the population. So B is incorrect. |
|
|
|
117 |
|
00:09:18,040 --> 00:09:25,240 |
|
Part C. It mentions here population, but total. So |
|
|
|
118 |
|
00:09:25,240 --> 00:09:30,300 |
|
this one is incorrect. Finally here, we are 97% |
|
|
|
119 |
|
00:09:30,300 --> 00:09:34,680 |
|
confident that the average total. So this one is |
|
|
|
120 |
|
00:09:34,680 --> 00:09:39,360 |
|
true of all. So here we have population and the |
|
|
|
121 |
|
00:09:39,360 --> 00:09:42,100 |
|
average of that population. So it makes sense that |
|
|
|
122 |
|
00:09:42,100 --> 00:09:43,260 |
|
this is the correct answer. |
|
|
|
123 |
|
00:09:46,520 --> 00:09:47,660 |
|
Number five. |
|
|
|
124 |
|
00:09:59,690 --> 00:10:03,290 |
|
Number five, confidence interval. Confidence |
|
|
|
125 |
|
00:10:03,290 --> 00:10:06,610 |
|
interval was used to estimate the proportion of |
|
|
|
126 |
|
00:10:06,610 --> 00:10:10,170 |
|
statistics students that are females. A random |
|
|
|
127 |
|
00:10:10,170 --> 00:10:16,170 |
|
sample of 72 statistics students generated the |
|
|
|
128 |
|
00:10:16,170 --> 00:10:22,970 |
|
following 90% confidence interval, 0.438 |
|
|
|
129 |
|
00:10:22,970 --> 00:10:28,150 |
|
and 0.640. |
|
|
|
130 |
|
00:10:28,510 --> 00:10:32,890 |
|
42, based on the interval above the population |
|
|
|
131 |
|
00:10:32,890 --> 00:10:38,230 |
|
proportion of females equals to 0.6. So here we |
|
|
|
132 |
|
00:10:38,230 --> 00:10:44,310 |
|
have confidence interval for the female proportion |
|
|
|
133 |
|
00:10:44,310 --> 00:10:52,990 |
|
ranges between 0.438 up to 0.642. Based on this |
|
|
|
134 |
|
00:10:52,990 --> 00:10:57,050 |
|
interval. Is the population proportion of females |
|
|
|
135 |
|
00:10:57,050 --> 00:10:58,770 |
|
equal 60%? |
|
|
|
136 |
|
00:11:03,410 --> 00:11:06,690 |
|
So here we have from this point all the way up to |
|
|
|
137 |
|
00:11:06,690 --> 00:11:10,610 |
|
0.6. Is the population proportion of females equal |
|
|
|
138 |
|
00:11:10,610 --> 00:11:16,250 |
|
to 0.6? No. The answer is no, but know what? |
|
|
|
139 |
|
00:11:16,850 --> 00:11:24,320 |
|
Number A. No, and we are 90% sure of it. No, the |
|
|
|
140 |
|
00:11:24,320 --> 00:11:31,200 |
|
proportion is 54.17. See, maybe 60% is a |
|
|
|
141 |
|
00:11:31,200 --> 00:11:33,760 |
|
believable value of population proportion based on |
|
|
|
142 |
|
00:11:33,760 --> 00:11:38,080 |
|
information about it. He said yes, and we are 90% |
|
|
|
143 |
|
00:11:38,080 --> 00:11:44,300 |
|
sure of it. So which one is correct? Farah. Which |
|
|
|
144 |
|
00:11:44,300 --> 00:11:44,900 |
|
one is correct? |
|
|
|
145 |
|
00:11:50,000 --> 00:11:56,760 |
|
B says the proportion is 54. 54 if we take the |
|
|
|
146 |
|
00:11:56,760 --> 00:12:01,120 |
|
average of these two values, the answer is 54. But |
|
|
|
147 |
|
00:12:01,120 --> 00:12:04,960 |
|
the true proportion is not the average of the two |
|
|
|
148 |
|
00:12:04,960 --> 00:12:07,640 |
|
endpoints. |
|
|
|
149 |
|
00:12:08,440 --> 00:12:09,500 |
|
So B is incorrect. |
|
|
|
150 |
|
00:12:12,780 --> 00:12:16,320 |
|
If you look at A, the answer is no. And we |
|
|
|
151 |
|
00:12:16,320 --> 00:12:20,440 |
|
mentioned before that this interval may Or may not |
|
|
|
152 |
|
00:12:20,440 --> 00:12:25,380 |
|
contains the true proportion, so A is incorrect. |
|
|
|
153 |
|
00:12:26,700 --> 00:12:32,640 |
|
Now C, maybe. So C is the correct statement, maybe |
|
|
|
154 |
|
00:12:32,640 --> 00:12:35,820 |
|
60% is believable value of the population |
|
|
|
155 |
|
00:12:35,820 --> 00:12:39,020 |
|
proportion based on the information about. So C is |
|
|
|
156 |
|
00:12:39,020 --> 00:12:44,440 |
|
the correct answer. A6, number six. |
|
|
|
157 |
|
00:12:48,590 --> 00:12:49,550 |
|
Number six. |
|
|
|
158 |
|
00:13:21,280 --> 00:13:23,800 |
|
So up to this point, we have the same information |
|
|
|
159 |
|
00:13:23,800 --> 00:13:27,440 |
|
for the previous problem. Using the information |
|
|
|
160 |
|
00:13:27,440 --> 00:13:31,440 |
|
about what total size sample would be necessary if |
|
|
|
161 |
|
00:13:31,440 --> 00:13:35,460 |
|
we wanted to estimate the true proportion within |
|
|
|
162 |
|
00:13:35,460 --> 00:13:43,620 |
|
minus positive or minus 0.108 using 95% |
|
|
|
163 |
|
00:13:43,620 --> 00:13:46,320 |
|
confidence. Now here we are looking for the sample |
|
|
|
164 |
|
00:13:46,320 --> 00:13:49,160 |
|
size that is required to estimate the true |
|
|
|
165 |
|
00:13:49,160 --> 00:13:53,720 |
|
proportion to be within 8% plus or minus 8% using |
|
|
|
166 |
|
00:13:53,720 --> 00:13:54,720 |
|
95% confidence. |
|
|
|
167 |
|
00:13:58,640 --> 00:14:05,360 |
|
The formula first, n equals z squared c plus one. |
|
|
|
168 |
|
00:14:08,740 --> 00:14:14,240 |
|
We have pi, one minus pi divided by e squared. |
|
|
|
169 |
|
00:14:15,740 --> 00:14:21,120 |
|
Now, pi is not given. So in this case either we |
|
|
|
170 |
|
00:14:21,120 --> 00:14:25,880 |
|
use a sinus sample in order to estimate the sample |
|
|
|
171 |
|
00:14:25,880 --> 00:14:30,400 |
|
proportion, Or use Pi to be 0.5. So in this case |
|
|
|
172 |
|
00:14:30,400 --> 00:14:35,900 |
|
we have to use Pi 1 half. If you remember last |
|
|
|
173 |
|
00:14:35,900 --> 00:14:39,720 |
|
time I gave you this equation. Z alpha over 2 |
|
|
|
174 |
|
00:14:39,720 --> 00:14:44,820 |
|
divided by 2 squared. So we have this equation. |
|
|
|
175 |
|
00:14:45,900 --> 00:14:49,280 |
|
Because Pi is not given, just use Pi to be 1 half. |
|
|
|
176 |
|
00:14:50,060 --> 00:14:54,880 |
|
Or you may use this equation. shortcut formula. In |
|
|
|
177 |
|
00:14:54,880 --> 00:15:02,120 |
|
this case, here we are talking about 95%. So |
|
|
|
178 |
|
00:15:02,120 --> 00:15:07,240 |
|
what's the value of Z? 196. 2 times E. |
|
|
|
179 |
|
00:15:10,100 |
|
E is 8%. So 196 divided by 2 times E, the quantity |
|
|
|
180 |
|
00:15:17,140 |
|
squared. Now the answer of this problem 150. So |
|
|
|
181 |
|
00:15:25,540 |
|
approximately 150. |
|
|
|
182 |
|
00:15:29,160 |
|
So 150 is the correct answer. So again, here we |
|
|
|
183 |
|
00:15:33,520 |
|
used pi to be 1 half because P is not given. And |
|
|
|
184 |
|
00:15:41,460 |
|
simple calculation results in 150 for the sample |
|
|
|
185 |
|
00:15:46,580 |
|
size. So P is the correct answer, 7. |
|
|
|
186 |
|
00:15:55,220 |
|
Number seven. |
|
|
|
187 |
|
00:16:00,480 |
|
Number seven. When determining the sample size |
|
|
|
188 |
|
00:16:03,820 |
|
necessarily for estimating the true population |
|
|
|
189 |
|
00:16:05,820 |
|
mean, which factor is not considered when sampling |
|
|
|
190 |
|
00:16:10,560 |
|
with replacement? Now here, if you remember the |
|
|
|
191 |
|
00:16:14,960 |
|
formula for the sample size. |
|
|
|
192 |
|
00:16:38,820 |
|
Now, which factor is not considered when sampling |
|
|
|
193 |
|
00:16:43,460 |
|
without weight replacement? Now, the population |
|
|
|
194 |
|
00:16:47,120 |
|
size, the population size is not in this quantity, |
|
|
|
195 |
|
00:16:51,600 |
|
so A is the correct answer. B, the population |
|
|
|
196 |
|
00:16:54,420 |
|
standard deviation, sigma is here. C, the level of |
|
|
|
197 |
|
00:16:58,820 |
|
confidence desired in the estimate, we have Z. The |
|
|
|
198 |
|
00:17:03,090 |
|
allowable or tolerable seminar, we have it here. |
|
|
|
199 |
|
00:17:07,370 |
|
So eight is the correct answer. |
|
|
|
200 |
|
00:17:13,290 |
|
Eight. |
|
|
|
201 |
|
00:17:20,020 |
|
Supposedly, I'm supposed to focus on the companies |
|
|
|
202 |
|
00:17:22,600 --> 00:17:25,640 |
|
that you're working on now. It turns out to be one |
|
|
|
203 |
|
00:17:25,640 |
|
of them. I'm not sure if I'm on the right track. |
|
|
|
204 |
|
00:17:28,420 |
|
To make more use of it as a reference for the |
|
|
|
205 |
|
00:17:31,040 |
|
update. |
|
|
|
206 |
|
00:17:43,820 |
|
Now, which of the following will result in a |
|
|
|
207 |
|
00:17:47,240 |
|
reduced interval width? So here we are talking |
|
|
|
208 |
|
00:17:50,100 |
|
about reducing the width of the interval. Number |
|
|
|
209 |
|
00:17:55,580 |
|
one. Here, if you look carefully at this equation, |
|
|
|
210 |
|
00:17:59,040 |
|
increase the sample size, the error Z up over 2 |
|
|
|
211 |
|
00:18:08,140 |
|
sigma over n. So this is the error state. Now, |
|
|
|
212 |
|
00:18:17,240 |
|
based on this equation, if we increase the sample |
|
|
|
213 |
|
00:18:20,360 |
|
size, the error will decrease. That means we |
|
|
|
214 |
|
00:18:25,180 |
|
reduce the interval with it. So A is the correct |
|
|
|
215 |
|
00:18:28,360 |
|
answer. Look at B. Increase the confidence level. |
|
|
|
216 |
|
00:18:34,030 |
|
Increasing the confidence level, it means increase |
|
|
|
217 |
|
00:18:36,770 |
|
Z, increase E, that means we have wider confidence |
|
|
|
218 |
|
00:18:41,030 |
|
interval, so B is incorrect. Increase the |
|
|
|
219 |
|
00:18:43,910 |
|
population mean, it doesn't matter actually, so |
|
|
|
220 |
|
00:18:46,790 --> 00:18:50,250 |
|
it's not correct. Increase the sample mean also. |
|
|
|
221 |
|
00:18:50,770 |
|
So C and D are incorrect totally, so B is |
|
|
|
222 |
|
00:18:54,990 |
|
incorrect, so E is the correct answer. So the |
|
|
|
223 |
|
00:18:57,670 |
|
correct answer is A, nine. |
|
|
|
224 |
|
00:19:07,140 |
|
A major department store chain is interested in |
|
|
|
225 |
|
00:19:10,500 |
|
estimating the average amount each credit and |
|
|
|
226 |
|
00:19:13,560 |
|
customers spent on their first visit to the |
|
|
|
227 |
|
00:19:16,560 |
|
chain's new store in the mall. 15 credit cards |
|
|
|
228 |
|
00:19:21,380 --> 00:19:26,540 |
|
accounts were randomly sampled and analyzed with |
|
|
|
229 |
|
00:19:26,540 --> 00:19:29,320 |
|
the following results. So here we have this |
|
|
|
230 |
|
00:19:29,320 --> 00:19:34,880 |
|
information about the 15 data points. We have x |
|
|
|
231 |
|
00:19:34,880 --> 00:19:35,220 |
|
bar. |
|
|
|
232 |
|
00:19:38,550 --> 00:19:42,150 |
|
of $50.5. |
|
|
|
233 |
|
00:19:43,470 --> 00:19:47,390 |
|
And S squared, the sample variance is 400. |
|
|
|
234 |
|
00:19:49,890 --> 00:19:52,750 |
|
Construct 95 confidence interval for the average |
|
|
|
235 |
|
00:19:52,750 --> 00:19:55,570 |
|
amount it's credit card customer spent on their |
|
|
|
236 |
|
00:19:55,570 |
|
first visit to the chain. It's a new store. It's |
|
|
|
237 |
|
00:20:01,230 |
|
in the mall, assuming the amount spent follows a |
|
|
|
238 |
|
00:20:04,310 |
|
normal distribution. |
|
|
|
239 |
|
00:20:08,090 |
|
In this case, we should use T instead of Z. So the |
|
|
|
240 |
|
00:20:13,150 |
|
formula should be X bar plus or minus T, alpha |
|
|
|
241 |
|
00:20:16,310 |
|
over 2S over root N. |
|
|
|
242 |
|
00:20:23,110 |
|
So X bar is 50.5. T, we should use the T table. In |
|
|
|
243 |
|
00:20:29,350 |
|
this case, here we are talking about 95%. |
|
|
|
244 |
|
00:20:36,830 |
|
So that means alpha is 5%, alpha over 2, 0, 2, 5. |
|
|
|
245 |
|
00:20:44,770 |
|
So now we are looking for 2, 0, 2, 5, and degrees |
|
|
|
246 |
|
00:20:48,930 |
|
of freedom. N is 15. It says that 15 credit cards. |
|
|
|
247 |
|
00:20:55,770 |
|
So 15 credit cards accounts for random samples. So |
|
|
|
248 |
|
00:20:59,110 |
|
N equals 15. So since N is 15, Degrees of freedom |
|
|
|
249 |
|
00:21:05,470 |
|
is 14. Now we may use the normal, the T table in |
|
|
|
250 |
|
00:21:09,850 |
|
order to find the value of T in |
|
|
|
251 |
|
00:21:16,250 |
|
the upper tier actually. So what's the value if |
|
|
|
252 |
|
00:21:19,270 --> 00:21:26,350 |
|
you have the table? So look at degrees of freedom |
|
|
|
253 |
|
00:21:26,350 --> 00:21:33,090 |
|
14 under the probability of 0 to 5. |
|
|
|
254 |
|
00:21:40,190 --> 00:21:45,050 |
|
So again, we are looking for degrees of freedom |
|
|
|
255 |
|
00:21:45,050 --> 00:21:49,170 |
|
equal 14 under 2.5%. |
|
|
|
256 |
|
00:22:04,850 --> 00:22:11,390 |
|
0.5 plus or minus 2 |
|
|
|
257 |
|
00:22:11,390 --> 00:22:18,390 |
|
.1448. S squared is given 400. Take square root of |
|
|
|
258 |
|
00:22:18,390 --> 00:22:25,570 |
|
this quantity 20 over root n over root 15. And the |
|
|
|
259 |
|
00:22:25,570 --> 00:22:30,350 |
|
answer, just simple calculation will give |
|
|
|
260 |
|
00:22:34,250 --> 00:22:38,410 |
|
This result, so D is the correct answer. So the |
|
|
|
261 |
|
00:22:38,410 --> 00:22:45,870 |
|
answer should be 50.5 plus or minus 11.08. So D is |
|
|
|
262 |
|
00:22:45,870 --> 00:22:49,170 |
|
the correct answer. So this one is straightforward |
|
|
|
263 |
|
00:22:49,170 --> 00:22:52,990 |
|
calculation, gives part D to be the correct |
|
|
|
264 |
|
00:22:52,990 --> 00:22:55,750 |
|
answer. Any question? |
|
|
|
265 |
|
00:22:58,510 --> 00:23:00,110 |
|
11, 10? |
|
|
|
266 |
|
00:23:03,110 --> 00:23:07,250 |
|
Private colleges and universities rely on money |
|
|
|
267 |
|
00:23:07,250 --> 00:23:12,730 |
|
contributed by individuals and corporations for |
|
|
|
268 |
|
00:23:12,730 --> 00:23:17,950 |
|
their operating expenses. Much of this money is |
|
|
|
269 |
|
00:23:17,950 --> 00:23:24,090 |
|
put into a fund called an endowment, and the |
|
|
|
270 |
|
00:23:24,090 --> 00:23:27,530 |
|
college spends only the interest earned by the |
|
|
|
271 |
|
00:23:27,530 --> 00:23:33,130 |
|
fund. Now, here we have a recent It says that a |
|
|
|
272 |
|
00:23:33,130 --> 00:23:35,310 |
|
recent survey of eight private colleges in the |
|
|
|
273 |
|
00:23:35,310 --> 00:23:39,450 |
|
United States revealed the following endowment in |
|
|
|
274 |
|
00:23:39,450 --> 00:23:44,350 |
|
millions of dollars. So we have this data. So it's |
|
|
|
275 |
|
00:23:44,350 |
|
raw data. Summary statistics yield export to be |
|
|
|
276 |
|
00:23:50,650 |
|
180. |
|
|
|
277 |
|
00:23:57,010 |
|
So export. |
|
|
|
278 |
|
00:24:07,070 |
|
Now if this information is not given, you have to |
|
|
|
279 |
|
00:24:12,130 |
|
compute the average and standard deviation by the |
|
|
|
280 |
|
00:24:15,170 |
|
equations we know. But here, the mean and standard |
|
|
|
281 |
|
00:24:19,860 |
|
deviation are given. So just use this information |
|
|
|
282 |
|
00:24:23,200 |
|
anyway. Calculate 95 confidence interval for the |
|
|
|
283 |
|
00:24:27,480 |
|
mean endowment of all private colleges in the |
|
|
|
284 |
|
00:24:30,140 |
|
United States, assuming a normal distribution for |
|
|
|
285 |
|
00:24:34,520 |
|
the endowment. Here we have 95%. |
|
|
|
286 |
|
00:24:39,300 |
|
Now |
|
|
|
287 |
|
00:24:42,600 |
|
what's the sample size? It says that eight. So N |
|
|
|
288 |
|
00:24:48,480 --> 00:24:53,900 |
|
is eight. So again, simple calculation. So |
|
|
|
289 |
|
00:24:53,900 --> 00:24:59,680 |
|
explore, plus or minus T, S over root N. So use |
|
|
|
290 |
|
00:24:59,680 --> 00:25:04,200 |
|
the same idea for the previous one. And the answer |
|
|
|
291 |
|
00:25:04,200 --> 00:25:10,420 |
|
for number 10 is part D. So D is the correct |
|
|
|
292 |
|
00:25:10,420 --> 00:25:17,380 |
|
answer. So again, For eleven, D is the correct |
|
|
|
293 |
|
00:25:17,380 --> 00:25:22,680 |
|
answer. For ten, D is the correct answer. Next. So |
|
|
|
294 |
|
00:25:22,680 --> 00:25:26,280 |
|
this one is similar to the one we just did. |
|
|
|
295 |
|
00:25:30,660 --> 00:25:31,260 |
|
Eleven. |
|
|
|
296 |
|
00:25:47,140 --> 00:25:51,140 |
|
Here it says that rather than examine the records |
|
|
|
297 |
|
00:25:51,140 --> 00:25:56,220 |
|
of all students, the dean took a random sample of |
|
|
|
298 |
|
00:25:56,220 --> 00:26:01,340 |
|
size 200. So we have large university. Here we |
|
|
|
299 |
|
00:26:01,340 --> 00:26:04,860 |
|
took representative sample of size 200. |
|
|
|
300 |
|
00:26:26,980 --> 00:26:31,900 |
|
How many students would be to be assembled? It |
|
|
|
301 |
|
00:26:31,900 --> 00:26:34,540 |
|
says that if the dean wanted to estimate the |
|
|
|
302 |
|
00:26:34,540 --> 00:26:38,040 |
|
proportion of all students, The saving financial |
|
|
|
303 |
|
00:26:38,040 --> 00:26:46,100 |
|
aid to within 3% with 99% probability. How many |
|
|
|
304 |
|
00:26:46,100 --> 00:26:51,620 |
|
students would need to be sampled? So we have the |
|
|
|
305 |
|
00:26:51,620 --> 00:26:56,920 |
|
formula, if you remember, n equals z y 1 minus y |
|
|
|
306 |
|
00:26:56,920 --> 00:27:00,860 |
|
divided by e. So we have z squared. |
|
|
|
307 |
|
00:27:03,640 --> 00:27:09,200 |
|
Now, y is not given. If Pi is not given, we have |
|
|
|
308 |
|
00:27:09,200 --> 00:27:14,180 |
|
to look at either B or 0.5. Now in this problem, |
|
|
|
309 |
|
00:27:15,000 --> 00:27:18,900 |
|
it says that Dean selected 200 students, and he |
|
|
|
310 |
|
00:27:18,900 --> 00:27:23,800 |
|
finds that out of this number, 118 of them are |
|
|
|
311 |
|
00:27:23,800 --> 00:27:26,480 |
|
receiving financial aid. So based on this |
|
|
|
312 |
|
00:27:26,480 --> 00:27:30,480 |
|
information, we can compute B. So B is x over n. |
|
|
|
313 |
|
00:27:30,700 --> 00:27:34,840 |
|
It's 118 divided by 200. And this one gives? |
|
|
|
314 |
|
00:27:41,090 |
|
So in this case, out of 200 students, 118 of them |
|
|
|
315 |
|
00:27:46,310 |
|
are receiving financial aid. That means the |
|
|
|
316 |
|
00:27:49,630 |
|
proportion, the sample proportion, is 118 divided |
|
|
|
317 |
|
00:27:53,730 |
|
by 200, which is 0.59. So we have to use this |
|
|
|
318 |
|
00:27:57,690 |
|
information instead of pi. So n equals, |
|
|
|
319 |
|
00:28:08,050 |
|
now it's about 99%. 2.85. Exactly, it's one of |
|
|
|
320 |
|
00:28:15,120 |
|
these. We have 2.57 and |
|
|
|
321 |
|
00:28:21,380 |
|
8. It says 99%. So |
|
|
|
322 |
|
00:28:30,220 |
|
here we have 99%. So what's left? |
|
|
|
323 |
|
00:28:42,180 --> 00:28:47,320 |
|
0.5 percent, this area. 0.5 to the right and 0.5 |
|
|
|
324 |
|
00:28:47,320 --> 00:28:52,500 |
|
to the left, so 005. Now if you look at 2.5 under |
|
|
|
325 |
|
00:28:52,500 --> 00:28:57,280 |
|
7, the answer is 0051, the other one 0049. |
|
|
|
326 |
|
00:28:59,840 --> 00:29:04,600 |
|
So either this one or the other value, so 2.57. or |
|
|
|
327 |
|
00:29:04,600 --> 00:29:07,600 |
|
2.58, it's better to take the average of these |
|
|
|
328 |
|
00:29:07,600 |
|
two. Because 005 lies exactly between these two |
|
|
|
329 |
|
00:29:13,320 |
|
values. So the score in this case, either 2.75 or |
|
|
|
330 |
|
00:29:20,780 |
|
2.78, or the average. And the exact one, 2.7, I'm |
|
|
|
331 |
|
00:29:30,880 --> 00:29:34,680 |
|
sorry, 2.576. The exact answer. |
|
|
|
332 |
|
00:29:38,000 --> 00:29:40,700 |
|
It's better to use the average if you don't |
|
|
|
333 |
|
00:29:40,700 --> 00:29:46,100 |
|
remember the exact answer. So it's the exact one. |
|
|
|
334 |
|
00:29:47,480 |
|
But 2.575 is okay. Now just use this equation, 2 |
|
|
|
335 |
|
00:29:53,440 |
|
.575 times square, times 59. |
|
|
|
336 |
|
00:30:03,900 |
|
1 minus 59 divided by the error. It's three |
|
|
|
337 |
|
00:30:09,440 --> 00:30:17,800 |
|
percent. So it's 0.0312 squared. So the answer in |
|
|
|
338 |
|
00:30:17,800 |
|
this case is part 2 |
|
|
|
339 |
|
00:30:23,420 |
|
.57 times 59 times 41 divided by 03 squared. The |
|
|
|
340 |
|
00:30:30,300 |
|
answer is part. |
|
|
|
341 |
|
00:30:41,650 |
|
You will get the exact answer if you use 2.576. |
|
|
|
342 |
|
00:30:48,190 |
|
You will get the exact answer. But anyway, if you |
|
|
|
343 |
|
00:30:51,230 |
|
use one of these, you will get approximate answer |
|
|
|
344 |
|
00:30:53,310 |
|
to be 1784. |
|
|
|
345 |
|
00:30:58,590 |
|
Any question? So in this case, we used the sample |
|
|
|
346 |
|
00:31:04,430 |
|
proportion instead of 0.5, because the dean |
|
|
|
347 |
|
00:31:11,240 |
|
selected a random sample of size 200, and he finds |
|
|
|
348 |
|
00:31:14,120 |
|
that 118 of them are receiving financial aid. That |
|
|
|
349 |
|
00:31:19,200 |
|
means the sample proportion is 118 divided by 200, |
|
|
|
350 |
|
00:31:25,360 |
|
which gives 0.59. So we have to use 59% as the |
|
|
|
351 |
|
00:31:30,420 |
|
sample proportion. Is it clear? Next, number |
|
|
|
352 |
|
00:31:38,360 |
|
three. |
|
|
|
353 |
|
00:31:41,700 |
|
An economist is interested in studying the incomes |
|
|
|
354 |
|
00:31:45,860 |
|
of consumers in a particular region. The |
|
|
|
355 |
|
00:31:51,620 |
|
population standard deviation is known to be 1 |
|
|
|
356 |
|
00:31:56,400 |
|
,000. A random sample of 50 individuals resulted |
|
|
|
357 |
|
00:32:00,560 |
|
in an average income of $15,000. What is the |
|
|
|
358 |
|
00:32:06,460 |
|
weight of the 90% confidence interval? So here in |
|
|
|
359 |
|
00:32:11,520 |
|
this example, the population standard deviation |
|
|
|
360 |
|
00:32:16,560 |
|
sigma is known. So sigma is $1000. |
|
|
|
361 |
|
00:32:24,600 |
|
Random sample of size 50 is selected. This sample |
|
|
|
362 |
|
00:32:32,280 |
|
gives an average of $15,000 ask |
|
|
|
363 |
|
00:32:41,430 |
|
about what is the width of the 90% confidence |
|
|
|
364 |
|
00:32:48,150 |
|
interval. So again, here |
|
|
|
365 |
|
00:32:55,630 |
|
we are asking about the width of the confidence |
|
|
|
366 |
|
00:32:58,710 |
|
interval. If we have a random sample of size 50, |
|
|
|
367 |
|
00:33:03,320 |
|
And that sample gives an average of $15,000. And |
|
|
|
368 |
|
00:33:07,560 |
|
we know that the population standard deviation is |
|
|
|
369 |
|
00:33:10,940 |
|
1,000. Now, what's the width of the 90% confidence |
|
|
|
370 |
|
00:33:17,580 --> 00:33:21,800 |
|
interval? Any idea of this? |
|
|
|
371 |
|
00:33:33,760 --> 00:33:40,020 |
|
So idea number one is fine. You may calculate the |
|
|
|
372 |
|
00:33:40,020 --> 00:33:43,400 |
|
lower limit and upper limit. And the difference |
|
|
|
373 |
|
00:33:43,400 --> 00:33:46,640 |
|
between these two gives the width. So idea number |
|
|
|
374 |
|
00:33:46,640 --> 00:33:51,360 |
|
one, the width equals the distance between upper |
|
|
|
375 |
|
00:33:51,360 --> 00:33:59,070 |
|
limit our limit minus lower limit. Now this |
|
|
|
376 |
|
00:33:59,070 --> 00:34:03,270 |
|
distance gives a width, that's correct. Let's see. |
|
|
|
377 |
|
00:34:04,710 --> 00:34:07,910 |
|
So in other words, you have to find the confidence |
|
|
|
378 |
|
00:34:07,910 --> 00:34:12,070 |
|
interval by using this equation x bar plus or |
|
|
|
379 |
|
00:34:12,070 --> 00:34:17,070 |
|
minus z sigma over root n, x bar is given. |
|
|
|
380 |
|
00:34:20,190 --> 00:34:28,690 |
|
Now for 90% we know that z equals 1.645 sigma is |
|
|
|
381 |
|
00:34:28,690 --> 00:34:32,670 |
|
1000 divided |
|
|
|
382 |
|
00:34:32,670 --> 00:34:40,850 |
|
by root 50 plus or minus. By calculator, 1000 |
|
|
|
383 |
|
00:34:40,850 --> 00:34:45,010 |
|
times this number divided by root 50 will give |
|
|
|
384 |
|
00:34:45,010 --> 00:34:49,190 |
|
around |
|
|
|
385 |
|
00:34:49,190 --> 00:34:50,730 |
|
232.6. |
|
|
|
386 |
|
00:34:58,290 --> 00:35:05,790 |
|
So the upper limit is this value and lower limit |
|
|
|
387 |
|
00:35:05,790 --> 00:35:09,650 |
|
147671. |
|
|
|
388 |
|
00:35:11,350 --> 00:35:14,250 |
|
So now the upper limit and lower limit are |
|
|
|
389 |
|
00:35:14,250 --> 00:35:18,590 |
|
computed. Now the difference between these two |
|
|
|
390 |
|
00:35:18,590 --> 00:35:24,010 |
|
values will give the weight. If you subtract these |
|
|
|
391 |
|
00:35:24,010 --> 00:35:26,030 |
|
two values, what equals 15,000? |
|
|
|
392 |
|
00:35:30,670 --> 00:35:37,190 |
|
And the answer is 465.13, around. |
|
|
|
393 |
|
00:35:40,050 --> 00:35:45,550 |
|
Maybe I took two minutes to figure the answer, the |
|
|
|
394 |
|
00:35:45,550 --> 00:35:49,350 |
|
right answer. But there is another one, another |
|
|
|
395 |
|
00:35:49,350 --> 00:35:52,790 |
|
idea, maybe shorter. It'll take shorter time. |
|
|
|
396 |
|
00:35:56,890 |
|
It's correct, but straightforward. Just compute |
|
|
|
397 |
|
00:36:00,230 --> 00:36:05,790 |
|
the lower and upper limits. And the width is the |
|
|
|
398 |
|
00:36:05,790 --> 00:36:07,190 |
|
difference between these two values. |
|
|
|
399 |
|
00:36:11,370 --> 00:36:16,050 |
|
If you look carefully at this equation, difference |
|
|
|
400 |
|
00:36:16,050 --> 00:36:21,560 |
|
between these two values gives the width. Now |
|
|
|
401 |
|
00:36:21,560 --> 00:36:25,880 |
|
let's imagine that the lower limit equals x bar |
|
|
|
402 |
|
00:36:25,880 |
|
minus |
|
|
|
403 |
|
00:36:28,920 |
|
the error term. And upper limit is also x bar plus |
|
|
|
404 |
|
00:36:36,340 |
|
the error term. |
|
|
|
405 |
|
00:36:41,460 |
|
Now if we add this, or if we subtract 2 from 1, |
|
|
|
406 |
|
00:36:47,900 |
|
you will get upper limit minus lower limit equals |
|
|
|
407 |
|
00:36:52,560 |
|
x |
|
|
|
408 |
|
00:36:55,740 |
|
bar cancels with 2x bar. If you subtract, w minus |
|
|
|
409 |
|
00:37:07,280 |
|
equals 2e. |
|
|
|
410 |
|
00:37:12,520 |
|
Upper limit minus lower limit is the width. So w, |
|
|
|
411 |
|
00:37:18,760 |
|
the width is twice the sampling error. So we have |
|
|
|
412 |
|
00:37:24,800 |
|
this new information, W equals twice of the margin |
|
|
|
413 |
|
00:37:29,980 |
|
of error. If we add 1 and 2, that will give lower |
|
|
|
414 |
|
00:37:36,400 |
|
limit plus upper limit equals to x bar. That means |
|
|
|
415 |
|
00:37:41,120 |
|
x bar equals lower limit plus upper limit divided |
|
|
|
416 |
|
00:37:45,800 |
|
by 2. |
|
|
|
417 |
|
00:37:53,970 |
|
the error, and X bar is the average of lower and |
|
|
|
418 |
|
00:37:59,790 |
|
upper limits. So by using this idea now, if we |
|
|
|
419 |
|
00:38:05,310 |
|
compute the error term, E equals Z sigma over root |
|
|
|
420 |
|
00:38:12,490 |
|
N, this quantity. |
|
|
|
421 |
|
00:38:17,350 |
|
And again, Z is 1645. Sigma is 1000. Divide by |
|
|
|
422 |
|
00:38:25,260 |
|
root 50. This gives 232.6. This is the error tier, |
|
|
|
423 |
|
00:38:36,300 |
|
or the margin of error. As we know, that's called |
|
|
|
424 |
|
00:38:40,040 --> 00:38:46,400 |
|
margin of error or sampling error. |
|
|
|
425 |
|
00:38:50,580 --> 00:38:56,190 |
|
So the error is this amount. The width is twice |
|
|
|
426 |
|
00:38:56,190 --> 00:39:04,490 |
|
this value. So W equals 2 times the error. And the |
|
|
|
427 |
|
00:39:04,490 --> 00:39:10,830 |
|
answer should be the same as the one we just... So |
|
|
|
428 |
|
00:39:10,830 --> 00:39:13,450 |
|
we end with the same result. Now which one is |
|
|
|
429 |
|
00:39:13,450 --> 00:39:17,370 |
|
shorter? Forget about my explanation up to this |
|
|
|
430 |
|
00:39:17,370 --> 00:39:22,570 |
|
point. We started from this one. We just computed |
|
|
|
431 |
|
00:39:22,570 --> 00:39:27,390 |
|
the error tier. I mean this amount. Then we found |
|
|
|
432 |
|
00:39:27,390 --> 00:39:32,950 |
|
the error to be this 232 multiply this by 2 will |
|
|
|
433 |
|
00:39:32,950 --> 00:39:35,550 |
|
give the sampling error or the, I'm sorry, will |
|
|
|
434 |
|
00:39:35,550 |
|
give the weight of the interval. Now imagine for |
|
|
|
435 |
|
00:39:39,630 |
|
this problem, the income, the average income is |
|
|
|
436 |
|
00:39:43,370 |
|
not given. |
|
|
|
437 |
|
00:39:47,590 |
|
Suppose x bar is not given. Now the question is, |
|
|
|
438 |
|
00:39:55,550 |
|
can you find the answer by using this idea? But |
|
|
|
439 |
|
00:40:00,030 |
|
here, without using x bar, we computed the |
|
|
|
440 |
|
00:40:04,390 |
|
sampling error to multiply this value by 2 and get |
|
|
|
441 |
|
00:40:07,810 |
|
the answer. |
|
|
|
442 |
|
00:40:11,810 |
|
So that's for number 12. |
|
|
|
443 |
|
00:40:16,830 --> 00:40:20,550 |
|
Again, for this particular example, there are two |
|
|
|
444 |
|
00:40:20,550 --> 00:40:25,610 |
|
ways to solve this problem. The first one, you |
|
|
|
445 |
|
00:40:25,610 --> 00:40:28,390 |
|
have to construct the confidence interval, then |
|
|
|
446 |
|
00:40:28,390 --> 00:40:32,910 |
|
subtract upper limit from the lower limit, you |
|
|
|
447 |
|
00:40:32,910 --> 00:40:38,030 |
|
will get the width of the interval. The other way, |
|
|
|
448 |
|
00:40:38,610 --> 00:40:42,150 |
|
just compute the error and multiply the answer by |
|
|
|
449 |
|
00:40:42,150 --> 00:40:48,210 |
|
2, you will get the same result. Number 13. |
|
|
|
450 |
|
00:40:56,020 --> 00:41:00,980 |
|
13th says that the head librarian at the Library |
|
|
|
451 |
|
00:41:00,980 --> 00:41:04,780 |
|
of Congress has asked her assistant for an |
|
|
|
452 |
|
00:41:04,780 --> 00:41:07,980 |
|
interval estimate of a mean number of books |
|
|
|
453 |
|
00:41:07,980 --> 00:41:12,720 |
|
checked out each day. The assistant provides the |
|
|
|
454 |
|
00:41:12,720 --> 00:41:23,000 |
|
following interval estimate. From 740 to 920 books |
|
|
|
455 |
|
00:41:23,000 --> 00:41:28,360 |
|
per day. If the head librarian knows that the |
|
|
|
456 |
|
00:41:28,360 --> 00:41:33,880 |
|
population standard deviation is 150 books shipped |
|
|
|
457 |
|
00:41:33,880 --> 00:41:37,420 |
|
outwardly, approximately how large a sample did |
|
|
|
458 |
|
00:41:37,420 --> 00:41:40,200 |
|
her assistant use to determine the interval |
|
|
|
459 |
|
00:41:40,200 --> 00:41:46,540 |
|
estimate? So the information we have is the |
|
|
|
460 |
|
00:41:46,540 --> 00:41:50,860 |
|
following. We have information about the |
|
|
|
461 |
|
00:41:50,860 --> 00:41:51,700 |
|
confidence interval. |
|
|
|
462 |
|
00:42:01,440 --> 00:42:02,800 |
|
920 books. |
|
|
|
463 |
|
00:42:05,940 --> 00:42:08,700 |
|
And sigma is known to be 150. |
|
|
|
464 |
|
00:42:12,980 --> 00:42:17,980 |
|
That's all we have. He asked about how large a |
|
|
|
465 |
|
00:42:17,980 |
|
sample did Herelsen's conclusion determine the |
|
|
|
466 |
|
00:42:20,880 --> 00:42:21,820 |
|
interval estimate. |
|
|
|
467 |
|
00:42:26,740 --> 00:42:31,850 |
|
Look at the answers. A is 2. B is 3, C is 12, it |
|
|
|
468 |
|
00:42:31,850 --> 00:42:33,950 |
|
cannot be determined from the information given. |
|
|
|
469 |
|
00:42:37,190 --> 00:42:43,190 |
|
Now, in order to find the number, the sample, we |
|
|
|
470 |
|
00:42:43,190 --> 00:42:48,350 |
|
need Sigma or E squared. Confidence is not given. |
|
|
|
471 |
|
00:42:50,550 --> 00:43:00,140 |
|
Sigma is okay. We can find the error. The error is |
|
|
|
472 |
|
00:43:00,140 --> 00:43:07,940 |
|
just W divided by 2. So the error is fine. I mean, |
|
|
|
473 |
|
00:43:08,100 --> 00:43:12,200 |
|
E is fine. E is B minus A or upper limit minus |
|
|
|
474 |
|
00:43:12,200 --> 00:43:17,100 |
|
lower limit divided by 2. So width divided by 2. |
|
|
|
475 |
|
00:43:17,240 --> 00:43:20,740 |
|
So this is fine. But you don't have information |
|
|
|
476 |
|
00:43:20,740 |
|
about Z. |
|
|
|
477 |
|
00:43:25,020 |
|
We are looking for N. So Z is not I mean, cannot |
|
|
|
478 |
|
00:43:29,550 |
|
be computed because the confidence level is not |
|
|
|
479 |
|
00:43:32,810 |
|
given. So the information is determined |
|
|
|
480 |
|
00:43:39,830 |
|
from the information given. Make sense? So we |
|
|
|
481 |
|
00:43:46,170 |
|
cannot compute this score. Z is fine. Z is 150. |
|
|
|
482 |
|
00:43:51,330 |
|
The margin of error, we can compute the margin by |
|
|
|
483 |
|
00:43:54,310 |
|
using this interval, the width. Divide by two |
|
|
|
484 |
|
00:43:59,090 |
|
gives the same result. Now for number 14, we have |
|
|
|
485 |
|
00:44:05,790 |
|
the same information. But here, |
|
|
|
486 |
|
00:44:14,450 |
|
she asked her assistant to use 25 days. So now, n |
|
|
|
487 |
|
00:44:22,030 |
|
is 25. We have the same information about the |
|
|
|
488 |
|
00:44:24,990 |
|
interval. |
|
|
|
489 |
|
00:44:32,020 |
|
And sigma is 150. |
|
|
|
490 |
|
00:44:36,300 |
|
So she asked her assistant to use 25 days of data |
|
|
|
491 |
|
00:44:40,800 |
|
to construct the interval estimate. So n is 25. |
|
|
|
492 |
|
00:44:44,980 |
|
What confidence level can she attach to the |
|
|
|
493 |
|
00:44:48,300 |
|
interval estimate? Now in this case, we are asking |
|
|
|
494 |
|
00:44:53,500 |
|
about confidence, not z. |
|
|
|
495 |
|
00:45:00,930 |
|
You have to distinguish between confidence level |
|
|
|
496 |
|
00:45:03,530 |
|
and z. We use z, I'm sorry, we use z level to |
|
|
|
497 |
|
00:45:08,130 --> 00:45:13,350 |
|
compute the z score. Now, which one is correct? 99 |
|
|
|
498 |
|
00:45:13,350 --> 00:45:21,670 |
|
.7, 99, 98, 95.4. Let's see. Now, what's the |
|
|
|
499 |
|
00:45:21,670 --> 00:45:25,070 |
|
average? I'm sorry, the formula is x bar plus or |
|
|
|
500 |
|
00:45:25,070 |
|
minus z sigma over root n. What's the average? In |
|
|
|
501 |
|
00:45:29,270 --> 00:45:34,710 |
|
this case, this is the formula we have. We are |
|
|
|
502 |
|
00:45:34,710 --> 00:45:38,770 |
|
looking about this one. Now, also there are two |
|
|
|
503 |
|
00:45:38,770 --> 00:45:43,250 |
|
ways to solve this problem. Either focus on the |
|
|
|
504 |
|
00:45:43,250 --> 00:45:47,950 |
|
aortia, or just find a continuous interval by |
|
|
|
505 |
|
00:45:47,950 --> 00:45:55,830 |
|
itself. So let's focus on this one. Z sigma over |
|
|
|
506 |
|
00:45:55,830 |
|
root of. |
|
|
|
507 |
|
00:45:59,620 |
|
And we have x bar. What's the value of x bar? x |
|
|
|
508 |
|
00:46:05,380 --> 00:46:15,240 |
|
bar 740 plus x |
|
|
|
509 |
|
00:46:15,240 --> 00:46:16,400 |
|
bar 830. |
|
|
|
510 |
|
00:46:25,380 --> 00:46:31,740 |
|
1660 divided by 2, 830. Now, z equals, I don't |
|
|
|
511 |
|
00:46:31,740 |
|
know, sigma, sigma is 150, n is 5. So here we have |
|
|
|
512 |
|
00:46:40,660 |
|
30 sigma. |
|
|
|
513 |
|
00:46:44,980 |
|
Now, what's the value of sigma? 36, so we have x |
|
|
|
514 |
|
00:46:51,560 --> 00:46:54,900 |
|
bar, now the value of x bar. |
|
|
|
515 |
|
00:47:02,330 --> 00:47:10,530 |
|
So we have x bar 830 plus or minus 30 there. |
|
|
|
516 |
|
00:47:15,290 --> 00:47:18,030 |
|
Now, if you look carefully at this equation, |
|
|
|
517 |
|
00:47:19,550 --> 00:47:24,570 |
|
what's the value of z in order to have this |
|
|
|
518 |
|
00:47:24,570 |
|
confidence interval, which is 740 and 920? |
|
|
|
519 |
|
00:47:36,170 |
|
So, Z should be... |
|
|
|
520 |
|
00:47:40,730 |
|
What's the value of Z? Now, 830 minus 3Z equals |
|
|
|
521 |
|
00:47:46,290 --> 00:47:46,870 |
|
this value. |
|
|
|
522 |
|
00:47:49,830 --> 00:47:53,390 |
|
So, Z equals... |
|
|
|
523 |
|
00:47:53,390 --> 00:47:56,450 |
|
3. |
|
|
|
524 |
|
00:47:56,830 --> 00:48:03,540 |
|
So, Z is 3. That's why. Now, Z is 3. What do you |
|
|
|
525 |
|
00:48:03,540 |
|
think the corresponding C level? |
|
|
|
526 |
|
00:48:11,460 |
|
99.7% If |
|
|
|
527 |
|
00:48:16,560 |
|
you remember for the 68 empirical rule 68, 95, 99 |
|
|
|
528 |
|
00:48:27,080 |
|
.7% In chapter 6 we said that 99.7% of the data |
|
|
|
529 |
|
00:48:33,760 |
|
falls within three standard deviations of the |
|
|
|
530 |
|
00:48:37,220 |
|
mean. So if these three, I am sure that we are |
|
|
|
531 |
|
00:48:41,980 |
|
using 99.7% for the confidence level. So for this |
|
|
|
532 |
|
00:48:50,340 |
|
particular example here, we have new information |
|
|
|
533 |
|
00:48:53,280 |
|
about the sample size. So N is 25. |
|
|
|
534 |
|
00:49:01,630 |
|
So just simple calculation x bar as I mentioned |
|
|
|
535 |
|
00:49:06,190 |
|
here. The average is lower limit plus upper limit |
|
|
|
536 |
|
00:49:11,510 |
|
divided by 2. So x bar equals 830. So now your |
|
|
|
537 |
|
00:49:18,270 |
|
confidence interval is x bar plus or minus z sigma |
|
|
|
538 |
|
00:49:25,130 |
|
over root n. z sigma over root n, z is unknown, |
|
|
|
539 |
|
00:49:32,190 |
|
sigma is 150, n is 25, which is 5, square root of |
|
|
|
540 |
|
00:49:37,030 |
|
it, so we'll have 3z. So now x bar 830 minus 3z, |
|
|
|
541 |
|
00:49:49,610 --> 00:49:53,870 |
|
this is the lower limit, upper limit 830 plus 3z. |
|
|
|
542 |
|
00:49:55,480 --> 00:49:59,000 |
|
Now, the confidence interval is given by 740 and |
|
|
|
543 |
|
00:49:59,000 --> 00:50:09,020 |
|
920. Just use the lower limit. 830 minus 3z equals |
|
|
|
544 |
|
00:50:09,020 --> 00:50:10,820 |
|
740. |
|
|
|
545 |
|
00:50:12,300 --> 00:50:18,280 |
|
Simple calculation here. 830 minus 740 is 90, |
|
|
|
546 |
|
00:50:18,660 --> 00:50:22,340 |
|
equals 3z. That means z equals 3. |
|
|
|
547 |
|
00:50:26,070 --> 00:50:29,750 |
|
Now the z value is 3, it means the confidence is |
|
|
|
548 |
|
00:50:29,750 --> 00:50:33,530 |
|
9917, so the correct answer is A. |
|
|
|
549 |
|
00:50:44,690 --> 00:50:49,390 |
|
The other way, you can use that one, by using the |
|
|
|
550 |
|
00:50:53,010 --> 00:50:55,830 |
|
Margin of error, which is the difference between |
|
|
|
551 |
|
00:50:55,830 --> 00:50:58,270 |
|
these two divided by two, you will get the same |
|
|
|
552 |
|
00:50:58,270 --> 00:51:02,630 |
|
result. So there are two methods, one of these |
|
|
|
553 |
|
00:51:02,630 --> 00:51:05,830 |
|
straightforward one. The other one, as you |
|
|
|
554 |
|
00:51:05,830 --> 00:51:13,550 |
|
mentioned, is the error term. It's B minus upper |
|
|
|
555 |
|
00:51:13,550 |
|
limit minus lower limit divided by two. Upper |
|
|
|
556 |
|
00:51:19,550 |
|
limit is 920. Minus 74 divided by 2. What's the |
|
|
|
557 |
|
00:51:27,450 --> 00:51:28,370 |
|
value for this one? |
|
|
|
558 |
|
00:51:34,570 --> 00:51:40,610 |
|
90. So the margin of error is 90. So E equals E. |
|
|
|
559 |
|
00:51:41,070 --> 00:51:43,790 |
|
Sigma or N equals? |
|
|
|
560 |
|
00:51:47,110 --> 00:51:50,810 |
|
All squared. So by using this equation you can get |
|
|
|
561 |
|
00:51:50,810 --> 00:51:56,860 |
|
your result. So, N is 25, Z is unknown, Sigma is |
|
|
|
562 |
|
00:51:56,860 --> 00:52:05,520 |
|
150, R is 90. This one squared. You will get the |
|
|
|
563 |
|
00:52:05,520 --> 00:52:10,020 |
|
same Z-score. Make sense? |
|
|
|
564 |
|
00:52:17,770 --> 00:52:21,810 |
|
Because if you take z to be three times one-fifth |
|
|
|
565 |
|
00:52:21,810 --> 00:52:25,150 |
|
divided by nine squared, you will get the same |
|
|
|
566 |
|
00:52:25,150 --> 00:52:30,790 |
|
result for z value. So both will give the same |
|
|
|
567 |
|
00:52:30,790 --> 00:52:35,790 |
|
result. So that's for the multiple choice |
|
|
|
568 |
|
00:52:35,790 |
|
problems. Any question? Let's move to the section |
|
|
|
569 |
|
00:52:42,430 --> 00:52:46,370 |
|
number two, true or false problems. |
|
|
|
570 |
|
00:52:47,810 --> 00:52:48,790 |
|
Number one, |
|
|
|
571 |
|
00:52:51,850 --> 00:52:57,950 |
|
a race car driver |
|
|
|
572 |
|
00:52:57,950 --> 00:53:03,670 |
|
tested his car for time from 0 to 60 mileage per |
|
|
|
573 |
|
00:53:03,670 --> 00:53:09,390 |
|
hour. And in 20 tests, obtained an average of 4.85 |
|
|
|
574 |
|
00:53:09,390 --> 00:53:16,660 |
|
seconds, with some deviation of 1.47 seconds. 95 |
|
|
|
575 |
|
00:53:16,660 --> 00:53:23,440 |
|
confidence interval for the 0 to 60 time is 4.62 |
|
|
|
576 |
|
00:53:23,440 --> 00:53:29,540 |
|
seconds up to 5.18. I think straightforward. Just |
|
|
|
577 |
|
00:53:29,540 --> 00:53:33,440 |
|
simple calculation, it will give the right answer. |
|
|
|
578 |
|
00:53:36,660 --> 00:53:40,640 |
|
x bar n, |
|
|
|
579 |
|
00:53:41,360 --> 00:53:43,620 |
|
so we have to use this equation. |
|
|
|
580 |
|
00:53:48,220 --> 00:53:54,020 |
|
You can do it. So it says the answer is false. You |
|
|
|
581 |
|
00:53:54,020 --> 00:53:58,340 |
|
have to check this result. So it's 4.5 plus or |
|
|
|
582 |
|
00:53:58,340 |
|
minus T. We have to find T. S is given to be 147 |
|
|
|
583 |
|
00:54:03,460 |
|
divided by root 20. Now, to find T, we have to use |
|
|
|
584 |
|
00:54:10,120 |
|
0 to 5 and 19. By this value here, you'll get the |
|
|
|
585 |
|
00:54:18,480 --> 00:54:22,160 |
|
exact answer. Part number two. |
|
|
|
586 |
|
00:54:24,980 --> 00:54:32,380 |
|
Given a sample mean of 2.1. So x bar is 2.1. |
|
|
|
587 |
|
00:54:33,680 --> 00:54:34,680 |
|
Excuse me? |
|
|
|
588 |
|
00:54:38,500 --> 00:54:45,920 |
|
Because n is small. Now, this sample, This sample |
|
|
|
589 |
|
00:54:45,920 --> 00:54:52,220 |
|
gives an average of 4.85, and standard deviation |
|
|
|
590 |
|
00:54:52,220 --> 00:55:02,420 |
|
based on this sample. So S, so X bar, 4.85, and S |
|
|
|
591 |
|
00:55:02,420 --> 00:55:09,640 |
|
is equal to 1.47. So this is not sigma, because it |
|
|
|
592 |
|
00:55:09,640 --> 00:55:15,210 |
|
says that 20 tests, so random sample is 20. This |
|
|
|
593 |
|
00:55:15,210 --> 00:55:19,390 |
|
sample gives an average of this amount and |
|
|
|
594 |
|
00:55:19,390 --> 00:55:21,350 |
|
standard deviation of this amount. |
|
|
|
595 |
|
00:55:29,710 --> 00:55:34,610 |
|
We are looking for the |
|
|
|
596 |
|
00:55:34,610 --> 00:55:40,470 |
|
continence interval, and we have two cases. First |
|
|
|
597 |
|
00:55:40,470 --> 00:55:43,630 |
|
case, if sigma is known, |
|
|
|
598 |
|
00:55:47,220 --> 00:55:48,240 |
|
Sigma is unknown. |
|
|
|
599 |
|
00:55:51,520 --> 00:55:58,440 |
|
Now for this example, sigma is unknown. So since |
|
|
|
600 |
|
00:55:58,440 --> 00:56:05,740 |
|
sigma is unknown, we have to use T distribution if |
|
|
|
601 |
|
00:56:05,740 --> 00:56:09,940 |
|
the distribution is normal or if N is large |
|
|
|
602 |
|
00:56:09,940 --> 00:56:14,380 |
|
enough. Now for this example, N is 20. So we have |
|
|
|
603 |
|
00:56:14,380 --> 00:56:17,860 |
|
to assume that the population is approximately |
|
|
|
604 |
|
00:56:17,860 --> 00:56:23,660 |
|
normal. So we have to use t. So my confidence |
|
|
|
605 |
|
00:56:23,660 --> 00:56:26,100 |
|
interval should be x bar plus or minus 3s over |
|
|
|
606 |
|
00:56:26,100 --> 00:56:32,560 |
|
root n. Now, number two. Given a sample mean of 2 |
|
|
|
607 |
|
00:56:32,560 --> 00:56:36,180 |
|
.1 and a population standard deviation. I |
|
|
|
608 |
|
00:56:36,180 --> 00:56:38,720 |
|
mentioned that population standard deviation is |
|
|
|
609 |
|
00:56:38,720 --> 00:56:43,900 |
|
given. So sigma is 0.7. So sigma is known in this |
|
|
|
610 |
|
00:56:43,900 --> 00:56:49,170 |
|
example. So in part two, sigma is given. Now, from |
|
|
|
611 |
|
00:56:49,170 --> 00:56:50,890 |
|
a sample of 10 data points, |
|
|
|
612 |
|
00:56:53,730 --> 00:56:56,190 |
|
we are looking for 90% confidence interval. |
|
|
|
613 |
|
00:56:58,790 --> 00:57:07,230 |
|
90% confidence interval will have a width of 2.36. |
|
|
|
614 |
|
00:57:16,460 --> 00:57:19,180 |
|
What is two times the assembling error? |
|
|
|
615 |
|
00:57:22,500 --> 00:57:28,040 |
|
So the answer is given. So the error here, error A |
|
|
|
616 |
|
00:57:28,040 --> 00:57:30,160 |
|
equals W. |
|
|
|
617 |
|
00:57:32,060 --> 00:57:34,120 |
|
So the answer is 1.16. |
|
|
|
618 |
|
00:57:40,520 --> 00:57:45,220 |
|
So he asked about given a sample, 90% confidence |
|
|
|
619 |
|
00:57:45,220 --> 00:57:50,540 |
|
interval will have a width of 2.36. Let's see if |
|
|
|
620 |
|
00:57:50,540 |
|
the exact width is 2.36 or not. Now we have x bar |
|
|
|
621 |
|
00:57:54,780 |
|
plus or minus z, sigma 1.8. x bar is 2.1, plus or |
|
|
|
622 |
|
00:58:03,240 |
|
minus. Now what's the error? 1.18. |
|
|
|
623 |
|
00:58:11,230 --> 00:58:16,370 |
|
this amount without calculation or you just use |
|
|
|
624 |
|
00:58:16,370 --> 00:58:19,590 |
|
this straightforward calculation here we are |
|
|
|
625 |
|
00:58:19,590 --> 00:58:23,530 |
|
talking about z about 90 percent so this amount 1 |
|
|
|
626 |
|
00:58:23,530 --> 00:58:30,330 |
|
.645 times sigma divided by root n for sure this |
|
|
|
627 |
|
00:58:30,330 --> 00:58:35,430 |
|
quantity equals 1.18 But you don't need to do that |
|
|
|
628 |
|
00:58:35,430 |
|
because the width is given to be 2.36. So E is 1 |
|
|
|
629 |
|
00:58:40,570 |
|
.18. So that saves time in order to compute the |
|
|
|
630 |
|
00:58:46,430 |
|
error term. So now 2.1 minus 1.8. 2.1 plus 1.8. |
|
|
|
631 |
|
00:58:56,350 |
|
That F, the width, is 2.36. |
|
|
|
632 |
|
00:59:02,010 |
|
that if the width equals this value. |
|
|
|
633 |
|
00:59:10,410 |
|
2.36. So I solved the problem if the width. But he |
|
|
|
634 |
|
00:59:15,270 |
|
asked about, do you know this value? I don't know |
|
|
|
635 |
|
00:59:18,430 --> 00:59:21,230 |
|
that one, so we have to compute the exact answer |
|
|
|
636 |
|
00:59:21,230 --> 00:59:28,230 |
|
now. So x bar 2.1 plus 1645 sigma |
|
|
|
637 |
|
00:59:34,480 --> 00:59:38,600 |
|
My calculator can find the error now. What's the |
|
|
|
638 |
|
00:59:38,600 |
|
value for this amount? My calculator. |
|
|
|
639 |
|
00:59:50,700 |
|
It's 5.75. 5.75. |
|
|
|
640 |
|
00:59:57,640 --> 01:00:01,830 |
|
So this is your error. So E equals this amount. So |
|
|
|
641 |
|
01:00:01,830 --> 01:00:05,370 |
|
W equals 2 plus 4. |
|
|
|
642 |
|
01:00:08,350 --> 01:00:17,050 |
|
So the error is 5.74. So what's the width? The |
|
|
|
643 |
|
01:00:17,050 |
|
width equals 2 times E. |
|
|
|
644 |
|
01:00:25,590 |
|
Again. This value, 1.645 times 1.7 divided by root |
|
|
|
645 |
|
01:00:30,340 |
|
10. |
|
|
|
646 |
|
01:00:37,280 |
|
Three point. |
|
|
|
647 |
|
01:00:44,020 |
|
So again, arrow is 3.64. So what's the width? |
|
|
|
648 |
|
01:00:51,160 --> 01:00:53,540 |
|
Twice this value, so two times this one. |
|
|
|
649 |
|
01:00:57,880 --> 01:00:59,560 |
|
7.28. |
|
|
|
650 |
|
01:01:02,120 --> 01:01:07,180 |
|
Now it says in the problem here we have width of 2 |
|
|
|
651 |
|
01:01:07,180 --> 01:01:09,200 |
|
.36. So it's incorrect. |
|
|
|
652 |
|
01:01:11,840 |
|
So just simple calculation gives width of 7.28, |
|
|
|
653 |
|
01:01:17,180 |
|
not 2.36. Number three. |
|
|
|
654 |
|
01:01:27,950 |
|
Look at number four. Other things be equal. As the |
|
|
|
655 |
|
01:01:32,850 |
|
confidence level for a confidence interval |
|
|
|
656 |
|
01:01:35,550 |
|
increases, the width of the interval increases. As |
|
|
|
657 |
|
01:01:41,250 |
|
the confidence level increases, confidence |
|
|
|
658 |
|
01:01:45,310 |
|
interval increases, the width of the interval |
|
|
|
659 |
|
01:01:47,650 |
|
increases. Correct. So that's true. Let's do |
|
|
|
660 |
|
01:01:52,750 |
|
number seven. |
|
|
|
661 |
|
01:01:56,840 |
|
A point estimate consists |
|
|
|
662 |
|
01:02:02,300 |
|
of a single sample statistic that is used to |
|
|
|
663 |
|
01:02:06,700 |
|
estimate the true population parameter. That's |
|
|
|
664 |
|
01:02:11,000 --> 01:02:15,940 |
|
correct because any point estimate, for example x |
|
|
|
665 |
|
01:02:15,940 --> 01:02:21,360 |
|
bar, is used to determine the confidence interval |
|
|
|
666 |
|
01:02:21,360 --> 01:02:25,600 |
|
for the unknown parameter mu. So a single |
|
|
|
667 |
|
01:02:25,600 --> 01:02:30,580 |
|
statistic can be used to estimate the true |
|
|
|
668 |
|
01:02:30,580 --> 01:02:33,400 |
|
population parameter, either X bar as a point |
|
|
|
669 |
|
01:02:33,400 --> 01:02:34,900 |
|
estimate or P. |
|
|
|
670 |
|
01:02:41,380 --> 01:02:48,000 |
|
So that's true. Number eight. The T distribution |
|
|
|
671 |
|
01:02:48,000 |
|
is used to develop a confidence interval estimate |
|
|
|
672 |
|
01:02:51,100 |
|
of the population mean when the population |
|
|
|
673 |
|
01:02:54,240 |
|
standard deviation is unknown. That's correct |
|
|
|
674 |
|
01:02:57,200 --> 01:03:01,240 |
|
because we are using T distribution if sigma is |
|
|
|
675 |
|
01:03:01,240 --> 01:03:03,740 |
|
not given and here we have to assume the |
|
|
|
676 |
|
01:03:03,740 --> 01:03:07,960 |
|
population is normal. 9. |
|
|
|
677 |
|
01:03:11,540 --> 01:03:15,180 |
|
The standardized normal distribution is used to |
|
|
|
678 |
|
01:03:15,180 --> 01:03:17,340 |
|
develop a confidence interval estimate of the |
|
|
|
679 |
|
01:03:17,340 --> 01:03:20,700 |
|
population proportion when the sample size is |
|
|
|
680 |
|
01:03:20,700 --> 01:03:22,820 |
|
large enough or sufficiently large. |
|
|
|
681 |
|
01:03:28,640 --> 01:03:32,640 |
|
The width |
|
|
|
682 |
|
01:03:32,640 --> 01:03:37,720 |
|
of a confidence interval equals twice the sampling |
|
|
|
683 |
|
01:03:37,720 --> 01:03:42,570 |
|
error. The weight equals twice the sample, so |
|
|
|
684 |
|
01:03:42,570 --> 01:03:46,370 |
|
that's true. A population parameter is used to |
|
|
|
685 |
|
01:03:46,370 |
|
estimate a confidence interval? No way. Because we |
|
|
|
686 |
|
01:03:50,650 |
|
use statistics to estimate the confidence |
|
|
|
687 |
|
01:03:53,570 |
|
interval. These are statistics. So we are using |
|
|
|
688 |
|
01:03:58,130 |
|
statistics to construct the confidence interval. |
|
|
|
689 |
|
01:04:04,190 |
|
Number 12. Holding the sample size fixed. In |
|
|
|
690 |
|
01:04:10,080 |
|
increasing level, the level of confidence in a |
|
|
|
691 |
|
01:04:14,560 |
|
confidence interval will necessarily lead to wider |
|
|
|
692 |
|
01:04:17,520 |
|
confidence interval. That's true. Because as C |
|
|
|
693 |
|
01:04:20,500 --> 01:04:24,840 |
|
level increases, Z becomes large, so we have large |
|
|
|
694 |
|
01:04:24,840 --> 01:04:29,670 |
|
width, so the confidence becomes wider. Last one, |
|
|
|
695 |
|
01:04:30,550 --> 01:04:33,150 |
|
holding the weight of a confidence interval fixed |
|
|
|
696 |
|
01:04:33,150 --> 01:04:36,190 |
|
and increasing the level of confidence can be |
|
|
|
697 |
|
01:04:36,190 --> 01:04:40,090 |
|
achieved with lower sample size with large sample |
|
|
|
698 |
|
01:04:40,090 --> 01:04:44,830 |
|
size. So it's false. So that's for section two. |
|
|
|
699 |
|
01:04:46,230 --> 01:04:49,970 |
|
One section is left, free response problems or |
|
|
|
700 |
|
01:04:49,970 --> 01:04:52,990 |
|
questions, you can do it at home. So next time, |
|
|
|
701 |
|
01:04:53,070 --> 01:04:57,530 |
|
inshallah, we'll start chapter nine. That's all. |
|
|
|
|