Lecture 15

Question 1

what allele causes the disease

Answer 1

the minor allele, since the majority of the population will have this allele

Question 2

what values will be in the beginning of the classification table?

Answer 2

with positive values

Question 3

positive + positive =

Answer 3

true positive

Question 4

negative + positive =

Answer 4

false negatives

Question 5

positive + negatives =

Answer 5

false positives

Question 6

negative + negative =

Answer 6

true negative

Question 7

define sensitivity (sn)

Answer 7

Proportion of people with the disease who have a positive test result

Question 8

define Specificity (Sp)

Answer 8

Proportion of people without the disease who have a negative test result.

Question 9

define Positive predictive value (PPV)

Answer 9

Proportion of people with a positive test who have the disease. PPV=TP/(TP+FP)

Question 10

define Negative predictive value (NPV)

Answer 10

Proportion of people with a negative test who do not have the disease NPV=TN/(TN+FN)

Question 11

positive likelihood ratio

Answer 11

How much more likely is a positive test to be found in a person with the disease than in a person without it?

Question 12

negative likelihood ratio

Answer 12

How much more likely is a negative test to be found in a person without the disease than in a person with it?

Question 13

what is the positive outcome

Answer 13

being ill

Question 14

what is negative outcome

Answer 14

not having the disease

Question 15

when the LR+ is high and the LR- are low, how good is the model?

Answer 15

the model is really good. LR+ higher than 1 is required for precise test

Question 16

define receiver operating characteristic (ROC) curve

Answer 16

A graph of sensitivity against 1

Question 17

what is the ideal ROC curve?

Answer 17

A perfect ROC curve will go from (1;0), to (1;1), it will have a right angle

Question 18

does multinomial regression needs larger sample size than binary logistic regression?

Answer 18

yes