Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05
-
Upload
derek-powers -
Category
Documents
-
view
35 -
download
0
description
Transcript of Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05
![Page 1: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/1.jpg)
1Week 3.
![Page 2: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/2.jpg)
2Week 3.
![Page 3: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/3.jpg)
3
“oil” = oil is present“+” = a test for oil is positive
“-” = a test for oil is negative
oil
no oil
+
+
false negativefalse positive
-
-
![Page 4: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/4.jpg)
4
P(oil) = 0.3P(+ | oil) = 0.9
P(+ | no oil) = 0.4
oil
no oil
+
+
-
-
P(oil) = 0.3
P(+ | oil) = 0.9 P(oil +)= (0.3)(0.9)= 0.27
![Page 5: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/5.jpg)
5
P(oil) = 0.3P(+ | oil) = 0.9
P(+ | no oil) = 0.4
0.3 oil
0.7 no oil
![Page 6: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/6.jpg)
6
P(oil) = 0.3P(+ | oil) = 0.9 P(- | oil) = 0.1
P(+ | no oil) = 0.4
oil
no oil
+
+
-
-
0.3
0.7
0.9
0.1
![Page 7: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/7.jpg)
7
P(oil) = 0.3P(+ | oil) = 0.9
P(+ | no oil) = 0.4
oil
no oil
+
+
-
-
0.3
0.7
0.9 0.27 oil+
0.4
0.6
0.03 oil-0.1
0.28 oil+
0.42 oil-
![Page 8: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/8.jpg)
8
oil
no oil
+
+
-
-
0.3
0.7
0.9 0.27
0.4
0.6
0.030.1
0.28
0.42
oil+
oil-oil+
oil-
S oil + 0.03 0.27 0.28 0.42
![Page 9: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/9.jpg)
9
oil
no oil
+
+
0.3
0.7
0.9 0.27
0.4 0.28
oil+
oil+
Oil contributes 0.27 to the total P(+) = 0.55.
![Page 10: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/10.jpg)
10
0.27
0.28
oil+
oil+
S oil + 0.03 0.27 0.28 0.42
Oil contributes 0.27 of the total P(+) = 0.27+0.28.
![Page 11: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/11.jpg)
11
+
+
-
-
0.01disease
0.98
0.03
The test for this infrequent disease seems to be reliable having only 3% false positives and 2% false negatives. What if we test positive?
0.99no disease
0.02
0.97
![Page 12: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/12.jpg)
12
+
+
-
-
0.01disease
0.98
0.03
We need to calculate P(diseased | +), the conditional probability that we have this disease GIVEN we’ve tested positive for it.
0.02
0.99no disease 0.97
0.0098
0.0002
0.0297
0.9603
![Page 13: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/13.jpg)
13
+
+
-
-
0.01disease
0.98
0.03
P(+) = 0.0098 + 0.0297 = 0.0395
P(disease | +) = P(disease+) / P(+)= .0098 / 0.0395 = 0.248.
0.02
0.99no disease 0.97
0.0098
0.0002
0.0297
0.9603
![Page 14: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/14.jpg)
14
+
+
-
-
0.01disease
0.98
0.03
EVEN FOR THIS ACCURATE TEST: P(diseased | +) is only around 25% because the non-diseased group is so predominant that most positives come from it.
0.02
0.99no disease 0.97
0.0098
0.0002
0.0297
0.9603
![Page 15: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/15.jpg)
15
FOR MEDICAL PRACTICE: Good diagnostictests will be of little use if the system is over-whelmed by lots of healthy people taking the test. Screen patients first.
FOR BUSINESS: Good sales people capably focus their efforts on likely buyers, leading to increased sales. They can be rendered ineffectiveby feeding them too many false leads, as with massive un-targeted sales promotions.
![Page 16: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/16.jpg)
16
probability 2 0.2 3 0.2 4 0.3 5 0.1 6 0.1 7 0.05 8 0.05 total 1
(3-17 of text)
![Page 17: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/17.jpg)
17
P(oil) = 0.3
oil
no oil
Cost to drill 130Reward for oil 400
0.3
0.7
net return “just drill”-130 + 400 = 270 drill oil drill no oil-130 + 000 = -130
A random variable is just a numerical function over the outcomes of a probability experiment.
![Page 18: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/18.jpg)
18
Definition of E XE X = sum of value times probability x p(x).
Key propertiesE(a X + b) = a E(X) + bE(X + Y) = E(X) + E(Y) (always, if such exist)
a. E(sum of 13 dice) = 13 E(one die) = 13(3.5).b. E(0.82 Ford US + Ford Germany - 20M) = 0.82 E(Ford US) + E(Ford Germany) - 20Mregardless of any possible dependence.
![Page 19: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/19.jpg)
19
probability product 2 1/36 2/36 3 2/36 6/36 4 3/36 12/36 5 4/36 20/36 6 5/36 30/36 7 6/36 42/36 8 5/36 40/36 9 4/36 36/36 10 3/36 30/36 11 2/36 22/36 12 1/36 12/36 sum 1 252/36 = 7
(3-15)of text
![Page 20: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/20.jpg)
20
probability product 2 0.2 0.4 3 0.2 0.6 4 0.3 1.2 5 0.1 0.5 6 0.1 0.6 7 0.05 0.35 8 0.05 0.4 total 1 4.05
(3-17 of text)
![Page 21: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/21.jpg)
21
oil
no oil
0.3
0.7
net return from policy“just drill.”-130 + 400 = 270 drill oil drill no-oil-130 + 0 = -130
E(X) = -10
Expected return from policy “just drill” is the probability weighted average (NET) returnE(NET) = (0.3) (270) + (0.7) (-130) = 81 - 91 = -10.
![Page 22: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/22.jpg)
22
oil
no oil
+
+
-
-
A test costing 20 is available. This test has: P(test + | oil) = 0.9 P(test + | no-oil) = 0.4.
0.27
0.03
0.28
0.30.90.1
0.40.6
0.7
0.42
“costs”TEST 20DRILL 130OIL 400
Is it worth 20 to test first?
![Page 23: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/23.jpg)
23
oil+ = -20 -130 + 400 = 250 0.27 67.5 oil- = -20 - 0 + 0 = - 20 .03 - 0.6 no oil+ = -20 -130 + 0 = -150 .28 - 42.0 no oil- = -20 - 0 + 0 = - 20 .42 - 8.4 total 1.00 16.5
E(NET) = .27 (250) - .03 (20) - .28 (150) - .42 (20) = 16.5 (for the “test first” policy). This average return is much preferred over the E(NET) = -10 of the “just drill” policy.
![Page 24: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/24.jpg)
24
x p(x) x p(x) x2 p(x) (x-4.05)2 p(x) 2 0.2 0.4 0.8 0.8405 3 0.2 0.6 1.8 0.2205 4 0.3 1.2 4.8 0.0005 5 0.1 0.5 2.5 0.09025 6 0.1 0.6 3.6 0.38025 7 0.05 0.35 2.45 0.435125 8 0.05 0.4 3.2 0.780125 total 1.00 4.05 19.15 2.7475quantity E X E X2 E (X - E X)2
terminology mean mean of squares variance = mean of sq dev
s.d. = root(2.7474) = root(19.15 - 4.052) = 1.6576
(3-17) of text
![Page 25: Raoul LePage Professor STATISTICS AND PROBABILITY stt.msu/~lepage click on STT315_Sp05](https://reader035.fdocuments.net/reader035/viewer/2022062721/568137e2550346895d9f896f/html5/thumbnails/25.jpg)
25