check_assumptions: Check assumptions of transitivity, consistency and...

In srhaile/mscpredmodel: Multiple Score Comparison of Prediction Models with Network Meta-Analysis

Description

Check assumptions of transitivity, consistency and homogeneity

Usage

check_transitivity(x, graph = FALSE)

check_homogeneity(object, dig = 3)

check_consistency(ps, mtype = c("consistency", "inconsistency")[1],
  ref = NULL)

Arguments

x	A set of computed (mscraw) or aggregated performance scores (mscagg)
graph	Should a graph of outcomes versus moderators be printed?
object	A consistency or inconsistency model, or the results of msc_network, msc_full, msc_direct, or msc_indirect.
dig	Number of digits to be printed, see format.pval.
ps	A set of raw performance estimates, from compute_performance
mtype	Type of model (default "consistency", else "inconsistency"). It is sufficient to write "c" or "i".
ref	Reference score all other scores should be compared to. (Default NULL, all pairwise comparisons are estimated)

Value

A tibble containing results of linear regression models of the effect size (outcome) against the cohort-specific average moderator value.

A vector containing tau-square, the Q likelihood ratio statistic, its degrees of freedom and its p-value.

A graph of the direct and indirect estimates for the various differences in score performance.

Functions

• check_transitivity: Check assumption of transitivity

• check_homogeneity: Check assumption of homogeneity

• check_consistency: Check assumption of consistency

Examples

dat <- msc_sample_data()
bssamp <- get_bs_samples(dat, id, study, outcome, n.samples = 8, 
                  scores = c("a", "b", "c", "d", "e", "f"), 
                  moderators = c("age", "female", "x1"))
perf <- compute_performance(bssamp, fn = calibration_slope, lbl = "CS")
agg <- aggregate_performance(perf)
check_transitivity(perf, graph = FALSE)
## Not run: check_transitivity(agg, graph = FALSE)
check_homogeneity(consistency(agg))
full <- msc_full(perf)
check_homogeneity(full)
check_consistency(perf)
## End(Not run)

srhaile/mscpredmodel documentation built on Sept. 13, 2019, 3:44 p.m.