This dataset was generated under conditions described in Leclerc et al.
(2015).Samples of n = 600 individuals from 120 families were generated: 40 families
of two parents and one child, 40 families of two parents and two children,
and 40 families of three generations (two grand-parents, four parents, and
two grandchildren). The coefficients of the block diagonal kinship matrix
were fixed at their expected theoretical values. The number of biallelic SNPs
was set to s = 50. The minor allele frequencies were randomly
sampled from Unif(0.001, 0.1). The genotypes of the 50 SNPs were
simulated assuming a linkage disequilibrium corresponding to a squared
correlation coefficient of r^2 = 0.5 between consecutive SNPs.
The two covariates follow Bernoulli(0.5) and Uniform(-0.2, 0.2) distributions
respectively. The polygenic heritability parameter was fixed at 0.5. Each
covariate parameter was set equal to 1 and the monotone increasing function
of the transformation model with censored data (Cheng et al., 1995) was fixed
at H(t) = log(t) in order to generate the survival traits. The censoring rate
was equal to 50%. The weight of each SNP was defined as the density function
of the Beta (1, 25) evaluated at the corresponding minor allele frequency.
The dataset includes simulated positions for the 50 SNPs, and the lower
and upper bounds of 4 sliding windows. Each window includes 10 SNPs,
overlapping with the previous and subsequent windows. A vector of size B*n
of permuted row indices is also included, where B=10,000. This is to
be used to compute the p-value of the test following the standard or matching
moments permutation approach.