Lab 9

pdf

School

University of British Columbia *

*We aren’t endorsed by this school

Course

125

Subject

Statistics

Date

Apr 3, 2024

Type

pdf

Pages

Uploaded by LieutenantFangKingfisher35

3/19/24, 3:39 PM Quiz:; E3 Assignment Lab Week9 E3 Assignment Lab Week9 Started: Mar 19 at 3:39p.m. Quiz Instructions This tutorial lab is about a dog named Marine. Marine was an 8 year old black Laborador dog trained for water rescue. But a group of researchers in Japan tested her to see how good she would be at identifying just by smell whether a patient had a particular kind of cancer or not (Sonada et al, GUT 2011). A movie about Marine is included in the course materials. Go to Canvas course website > media gallery > "Dog Cancer Detection". The researchers first trained Marine by having her smell what was in a bag from some patients with the cancer and in different bags from some other patients without the cancer but with some other ililnesses. Then they tested her on new bags to see whether she could correctly identify which one had the cancer. In one kind of test each bag contained the breath of a patient. One bag in each test contained breath from a patient with the cancer and 4 bags contained breath from patients without the cancer. In this case the test was repeated 36 times and Marine got it right 33 out of those 36 times. In the other kind of test stool samples from patients were in the bags. One bag in each test contained a stool sample from a patient with the cancer and 4 bags contained a stool sample from patients without the cancer. In this case the test was repeated 38 times and Marine got it right 37 out of those 38 times. This lab will concentrate on Marine's ability to identify the patient with cancer by smelling the breath of the patient. There are many possible research answers to the question about Marine’s ability to identify a patient with the cancer just by smell of their breath. They go all the way from Answer n=0%: Marine has an ability to correctly identify patients with the particular cancer by the smell of their breath 0% of the time. to Answer n=100%: Marine has an ability to correctly identify patients with the particular cancer by the smell of their breath 100% of the time. Each of these has a matching population observation saying how well Marine will perform over the long- run on the test. These go from PopnH n=0%: Marine correctly identifies patients with the particular cancer by the smell of their breath 0% of the time in the test setup over the long-run to PopnH n=100%: Marine correctly identifies patients with the particular cancer by the smell of their breath 100% of the time in the test setup over the long-run https://canvas.ubc.ca/courses/132235/quizzes/708110/take 1/5

3/19/24, 3:39 PM Quiz:; E3 Assignment Lab Week9 In this tutorial lab you will consider just one of these. The value of n you are to use is given to you in question 1 below. In this lab you have to modify and run a STATISTISTICS101 simulation program. The program is called MarineSIimSTATS101 and you can find in the canvas course website > modules > week9. You will be estimating some pvalue prediction probabilities for the Marine example. To complete the lab you will have to modify the simulation program to make Marine's long run success rate the ones you are assigned. The MarineSIimSTATS101 simulation program, creates a large population of Marine’s test performances where a long run population observation H (Marine's long run success rate) is made true. Each member of the population says whether Marine was successful in a test or unsuccessful. These scores are for 36 million tests of Marine of the kind done by Sonada et al. The program then extracts simple random samples (with replacement) of the same size as Sonada et al did -- that is, 36 tests -- but does this 100000 times and counts how many successful tests Marine had in the sample. You have to modify something about the program before running it. What you have to modify is in line 2 of the program. The value you give to xxx is how many of those tests in 36 million total tests of Marie are correct or successful based on the rate given to you in question 2 above. The value you give to yyy is how many of those tests in 36 million total tests of Marie are incorrect or unsuccessful. Be careful not to remove the hash signs when you change xxx and yyy. When you give a value to xxx and yyy use the long success rates expressed by "this corresponds to _ successes in every 36 tests over the long run" in questions 2 and 3 rather than the percentages. The percentages are approximations based on _ successes in every 36 tests. Question 1 0 pts Who are the two people working together on this assignment? Please include both their names and student numbers. If the two people working together get different values for Marine's long run correctness rate, the value that should be used is the value given to the partner with the lowest student number. Edit View Insert Format Tools Table 12pt v Paragraph v B 7 U A+v £ v T2v https://canvas.ubc.ca/courses/132235/quizzes/708110/take 2/5

3/19/24, 3:39 PM Quiz:; E3 Assignment Lab Week9 p @® oOowords <> Question 2 0 pts You should use the value 91.67% for the value of Marine's longrun correctness rate in this lab. This corresponds to 33 successes in every 36 tests over the long run. O OK. I got it. O | don't understand. I'll ask the TA. Question 3 14.5 pts Give this information for your assigned success rate. (1) Assigned success rate (as a %) (2) The value for xxx you put in the simulation program (0.5 points) (3) The value for yyy you put in the simulation program (0.5 points) (4) The total count out of 100000 you get from the first run of the program for the twosided pvalue. Don't convert this to a percentage /100000 (1 points) (5) The total count out of 100000 you get from the second run of the program for the twosided pvalue. Don't convert this to a percentage /100000 (1 points) (6) The total count out of 100000 you get from the third run of the program for the twosided pvalue. Don't convert this to a percentage /100000 (1 points) (7) The total count out of 100000 you get from the fourth run of the program for the twosided pvalue. Don't convert this to a percentage /100000 (1 points) (8) The total count out of 100000 you get from the fifth run of the program for the twosided pvalue. Don't convert this to a percentage /100000 (1 points) (9) The total count out of 100000 you get from the sixth run of the program for the twosided pvalue. Don't convert this to a percentage /100000 (1 points) https://canvas.ubc.ca/courses/132235/quizzes/708110/take 3/5

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

3/19/24, 3:39 PM Quiz:; E3 Assignment Lab Week9 (10) The total count out of 100000 you get from the seventh run of the program for the twosided pvalue. Don't convert this to a percentage /100000 (1 points) (11) The total count out of 100000 you get from the eighth run of the program for the twosided pvalue. Don't convert this to a percentage /100000 (1 points) (12) The total count out of 100000 you get from the ninth run of the program for the twosided pvalue. Don't convert this to a percentage /100000 (1 points) (13) The total count out of 100000 you get from the tenth run of the program for the twosided pvalue. Don't convert this to a percentage /100000 (1 points) (14) The highest count (out of 100000) out of all these in 10 runs / 100000 (0.5 points) (15) The lowest count (out of 100000) out of all these in 10 runs / 100000 (0.5 points) (16) Average of the numbers in (4) through to and including (13) out of 100000 / 100000 (1 points) (17) Based in question 14: The maximum estimated twosided pvalue for the long run success rate as a percentage (0.5 points). Recall the rule about reporting the results of a Monte Carlo estimate of a probability value. (18) Based in question 15: The minimum estimated twosided pvalue for the long run success rate as a percentage (0.5 points). Recall the rule about reporting the results of a Monte Carlo estimate of a probability value. (19) Based in question 16: The average estimated twosided pvalue for the long run success rate as a percentage (0.5 points). Recall the rule about reporting the results of a Monte Carlo estimate of a probability value. Edit View Insert Format Tools Table 12pt v Paragraph v B /7 U A v 2 v T2v https://canvas.ubc.ca/courses/132235/quizzes/708110/take 4/5

3/19/24, 3:39 PM Quiz:; E3 Assignment Lab Week9 p @® Oowords <> Not saved Submit Quiz https://canvas.ubc.ca/courses/132235/quizzes/708110/take 5/5

Related Documents

Homework-_5,6,7 .pdf

HST3501 Participant-Information-Letter_2022-2 (1).docx

Mgtsc312_F2023_Midterm1_Soln_21Oct2023.xlsx

SPED 5302 Week I Assignment.docx

Chapter 12 Exercises.docx

IMG_0660.png

IMG_5299.jpeg

Recommended textbooks for you

Glencoe Algebra 1, Student Edition, 9780079039897...

Algebra

ISBN:9780079039897

Author:Carter

Publisher:McGraw Hill

Big Ideas Math A Bridge To Success Algebra 1: Stu...

Algebra

ISBN:9781680331141

Author:HOUGHTON MIFFLIN HARCOURT

Publisher:Houghton Mifflin Harcourt

Holt Mcdougal Larson Pre-algebra: Student Edition...

Algebra

ISBN:9780547587776

Author:HOLT MCDOUGAL

Publisher:HOLT MCDOUGAL

Algebra: Structure And Method, Book 1

Algebra

ISBN:9780395977224

Author:Richard G. Brown, Mary P. Dolciani, Robert H. Sorgenfrey, William L. Cole

Publisher:McDougal Littell

College Algebra (MindTap Course List)

Algebra

ISBN:9781305652231

Author:R. David Gustafson, Jeff Hughes

Publisher:Cengage Learning

Mathematics For Machine Technology

Advanced Math

ISBN:9781337798310

Author:Peterson, John.

Publisher:Cengage Learning,

SEE MORE TEXTBOOKS

Recommended textbooks for you

Glencoe Algebra 1, Student Edition, 9780079039897...
Algebra
ISBN:9780079039897
Author:Carter
Publisher:McGraw Hill
Big Ideas Math A Bridge To Success Algebra 1: Stu...
Algebra
ISBN:9781680331141
Author:HOUGHTON MIFFLIN HARCOURT
Publisher:Houghton Mifflin Harcourt
Holt Mcdougal Larson Pre-algebra: Student Edition...
Algebra
ISBN:9780547587776
Author:HOLT MCDOUGAL
Publisher:HOLT MCDOUGAL
Algebra: Structure And Method, Book 1
Algebra
ISBN:9780395977224
Author:Richard G. Brown, Mary P. Dolciani, Robert H. Sorgenfrey, William L. Cole
Publisher:McDougal Littell
College Algebra (MindTap Course List)
Algebra
ISBN:9781305652231
Author:R. David Gustafson, Jeff Hughes
Publisher:Cengage Learning
Mathematics For Machine Technology
Advanced Math
ISBN:9781337798310
Author:Peterson, John.
Publisher:Cengage Learning,