util: Various utility functions

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

Description

Various utility functions

Usage

contrmat(trt1, trt2, ref, sc = NULL)

get_diff(x, s, ref)

get_ref(x, s, ref)

get_scores(x, s)

get_design(x, dl, s)

get_est(x, s)

get_var(x, s)

get_direct(x, s1, s2)

get_indirect(x, s1, s2, sc)

check_combn(x, to.check)

Arguments

trt1	vector of the 1st treatment / score in the contrast
trt2	vector of the 2nd treatment / score in the contrast
ref	The name of the reference treatment / score (character)
sc	A vector containing the full list of scores considered, so that the scores are not put into alphabetical order. This keeps the order of the scores in later network meta-analysis models the same as in other places.
x	A variety of possible inputs, depends on the function
s	vector of scores to be kept
dl	design.levels
s1	name of the 1st score in the contrast
s2	name of the 2nd score in the contrast
to.check	Name of score to be checked

Details

This function has been adapted slightly from the supplementary material of Law et al 2016:

Value

A design matrix

A new dataset with differences calculated

The number of the reference score

The names of the observed scores

A string of the design (combination of scores) of the cohort

a vector of performance estimates

a variance matrix

a data.frame

a data.frame

TRUE/FALSE

Functions

• contrmat: Make design matrix

• get_diff: Calculate differences between performance measures

• get_ref: Get reference score

• get_scores: Get non-missing scores

• get_design: Get design of the cohort

• get_est: Get estimated performance

• get_var: Get variance estimated performance

• get_direct: Find direct (head-to-head) comparisons

• get_indirect: Find indirect comparisons

• check_combn: Check if score is still in any of the cohorts?

See Also

Law, M.; Jackson, D.; Turner, R.; Rhodes, K. & Viechtbauer, W. Two new methods to fit models for network meta-analysis with random inconsistency effects BMC Medical Research Methodology, 2016, 16, 87.

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