Compare the fit of two qpgraph models — compare

Takes two data frames with model fits computed on two graphs for on the same populations and tests whether the scores of one graph are significantly better than the scores of the other.

compare_fits4(fit1, fit2, f2_blocks, f2_blocks_test, boot = FALSE, seed = NULL)

Arguments

fit1: The fit of the first graph
fit2: The fit of the second graph
f2_blocks: f2 blocks used for fitting fit1 and fit2. Used in combination with f2_blocks_test to compute f-statistics covariance matrix.
f2_blocks_test: f2 blocks which were not used for fitting fit1 and fit2
boot: If FALSE (the default), block-jackknife resampling will be used to compute standard errors. Otherwise, block-bootstrap resampling will be used to compute standard errors. If boot is an integer, that number will specify the number of bootstrap resamplings. If boot = TRUE, the number of bootstrap resamplings will be equal to the number of SNP blocks. If bootstrap resampling is enabled, empirical p-values (p_emp) and 95 confidence intervals (ci_low and ci_high) will be reported.
seed: Random seed used if boot is TRUE. Does not need to match a seed used in fitting the models

Examples

if (FALSE) {
nblocks = dim(example_f2_blocks)[3]
train = sample(1:nblocks, round(nblocks/2))
fit1 = qpgraph(example_f2_blocks[,,train], graph1)
fit2 = qpgraph(example_f2_blocks[,,train], graph2)
compare_fits4(fit1, fit2, example_f2_blocks[,,train], example_f2_blocks[,,-train])
}