fam.recrisk (version 0.1)

recrisk.mixture: Recurrence Risk based on a mixure of low and high risk families

Description

Recurrence risk assuming a mixture of low and high risk families based on a mixture of truncated binomial densities

Usage

recrisk.mixture(s, a, k, max.iter=1e4, eps=1e-6)

Arguments

s

vector of sizes of families

a

vector of number of affected members per family; same length as vector s

k

number of affecteds to condition on; the ascertainment criterion, e.g., at least k affecteds

max.iter

maximum number of EM iterations

eps

EM stop criterion value

Value

A list with the following values

recrisk.high

disease recurrence risk for high-risk group

recrisk.low

disease recurrence risk for low-risk group

prob.high.group

the mixture probability that a familiy is in the high-risk group (a scalar value)

prob.fam.high

a vector of posterior probabilities that families are in the high-risk group (same length as vector s)

iter

Number of EM iterations

Details

Model heterogeneity of risk across families by a mixture of truncated binomial distributions, assuming two groups: high and low risk.

References

Schaid DJ, McDonnell SK, Thibodeau SN. Familial recurrence risk with varying amount of family history. Under review at Genetic Epidemiology, 2018.

Examples

# NOT RUN {
s <- c(4,  8,  6,  6,  10, 8,  7,  8,  5,  6,  10, 4,  4,  8,  6,  8,  4,  5,  9, 
       9,  5,  4,  4,  7,  3,  3,  9,  5,  3,  3,  4,  6,  8,  3,  5,  8,  6,  8, 9,  3)
a <- c(3, 4, 2, 5, 4, 5, 3, 4, 3, 3, 8, 2, 0, 3, 4, 5, 2, 3, 4, 2, 1, 1,
       1, 1, 1, 0, 0, 0, 0, 0, 1, 0, 2, 0, 0, 1, 0, 2, 3, 0)
recrisk.mixture(s,a,k=1)
# }