Lecture 16

Question 1

for what survival analysis models are important

Answer 1

Engineering, Insurance, Marketing, Medicine, and many more application areas

Question 2

define censoring

Answer 2

Censoring is a form of missing data problem in which time to event is not observed for reasons such as termination of study before all recruited subjects have shown the event of interest or the subject has left the study prior to experiencing an event.

Question 3

how many variables we must have in a dataset?

Answer 3

more than 2

Question 4

define right censoring

Answer 4

true survival time is equal to or greater than observed survival time

Question 5

define left censoring

Answer 5

true survival time is less than or equal to the observed survival time

Question 6

define interval-censoring

Answer 6

true survival time is within a known time interval

Question 7

define the survival function

Answer 7

the probability of surviving at least to time t. where p1 is the proportion surviving the first period, p2 is the proportion surviving beyond the second period conditional on having survived up to the second period, and so on.

Question 8

define the proportion surviving period

Answer 8

is having survived up to period i is given by: (where ri is the number alive at the beginning of the period and di the number of deaths within the period. )

Question 9

define kaplan meier

Answer 9

is a non-parametric statistic used to estimate the survival function from lifetime data.

Question 10

define hazard

Answer 10

is the probability of dying (or experiencing the event in question) given that patients have survived up to a given point in time, or the risk for death at that moment.

Question 11

are there any assumptions made about the probability distribution of the hazard in Cox’s model?

Answer 11

No

Question 12

does hazard ratio depends on time?

Answer 12

no

Question 13

if the hazard ration is less than 1 then the predictor is

Answer 13

protective (i.e., associated with improved survival)

Question 14

if the hazard ration is more than 1 then the predictor is

Answer 14

then the predictor is associated with increased risk (or decreased survival)

Question 15

if hazard ratio for a predictor is close to 0 then

Answer 15

that predictor does not affect survival.

Question 16

what produces hazard ratio?

Answer 16

The antilog of an estimated regression coefficient, exp(bi )

Question 17

a p-value less than 0.05 indicates what

Answer 17

a violation of the proportionality assumption

Question 18

what the additive model estimates?

Answer 18

he difference in hazards: the change in hazard function due to the exposure of interest or stated more simply the absolute difference in the instantaneous failure rate per unit of change in the exposure variable.