Semi-parametric Survival Modelling Flashcards

1
Q

What is the cox-proportional hazard model?

A

h(t, x) = h0(t) * ψ(x ; B)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are the three choices for ψ?

A

exp( - B^(T) * x ) = log linear
1 +B^(T) * x = linear
ln( 1 + exp( - B^(T) * x ) ) = logisitic

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

How do you find the joint probability function?

A

product [ ψ(ij) / sum(k in R(tj)) ψ(k) ]

R(tj) is the risk set at time tj

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Given two groups, what is ψ(x) given by?

A

1 if individual is in group 1

exp(β) if individual is in group 2

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Give the likelihood for β given two groups

A

product ( exp( βxi ) / sum(k in R(tj) exp(βxj) )

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Formula for time dependent covariates?

A

h(t) = h0(t) * exp( β * x(t) )

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Formula for time dependent coefficients?

A

β(t) = ax + blog(t) * x

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is the difference between time dependent coefficients and covariates?

A

Coefficient means β’s change with time, covariates mean x’s change with time

How well did you know this?
1
Not at all
2
3
4
5
Perfectly