Generate tibble of covariate relations to test

The goal of NMproject's covariate modelling functions is to provide a stepwise covariate method with manual decision making. This important to ensure that the full model selection/evaluation criteria (should be defined in statistical analysis plans) can be applied at every step rather than just log likelihood ratio testing, where the most significant model may be unstable, may worsen model predictions or may only be slightly more significant than a more physiologically plausible covariate relationship.

The functions test_relations(), covariate_step_tibble(), bind_covariate_results() together comprise NMproject stepwise covariate method with manual decision. The goal is to be part way between PsN's SCM and completely manual process at each forward and backward elimination step. The syntax of how covariates are included is the same as PsN's SCM routine - See PsN documentation for more information.

test_relations(dtest, param, cov, state, continuous)

Arguments

dtest

Optional existing dtest to append (from an previous use test_relations()).

param

Character. Name of parameter(s).

cov

Character. Name of covariate(s).

state

Numeric or character. Number/name of state (see details).

continuous

Logical (default = TRUE). If FALSE, will treat the covariate as categorical.

Value

A tibble describing relationships to test.

Details

Setting vector values for param, cov, and state, will expand the grid to test each value with every other value greedily. This is similar to expand.grid() available states (see also add_cov()):

"2" or "linear"

PARCOV= ( 1 + THETA(1)*(COV -median))

"3" or "hockey-stick"

IF(COV.LE.median) PARCOV = ( 1 + THETA(1)*(COV - median)) IF(COV.GT.median) PARCOV = ( 1 + THETA(2)*(COV - median))

"4" or "exponential"

PARCOV= EXP(THETA(1)*(COV - median))

"5" or "power"

PARCOV= ((COV/median)**THETA(1))

"power1"

PARCOV= ((COV/median))

"power0.75"

PARCOV= ((COV/median)**0.75)

"6" or "log-linear"

PARCOV= ( 1 + THETA(1)*(LOG(COV) - log(median)))

See also

add_cov(), covariate_step_tibble(), bind_covariate_results()

Examples

dtest <- test_relations(param = c("KA", "K", "V"),
                        cov = c("LIN1", "LIN2", "LIN3", "RND1", "RND2", "RND3"),
                        state = c("linear", "power"),
                        continuous = TRUE) %>%
         test_relations(param = c("KA", "K", "V"),
                        cov = "BN1",
                        state = "linear",
                        continuous = FALSE)

dtest
#> # A tibble: 39 × 4
#>    param cov   state  continuous
#>    <chr> <chr> <chr>  <lgl>     
#>  1 KA    LIN1  linear TRUE      
#>  2 K     LIN1  linear TRUE      
#>  3 V     LIN1  linear TRUE      
#>  4 KA    LIN2  linear TRUE      
#>  5 K     LIN2  linear TRUE      
#>  6 V     LIN2  linear TRUE      
#>  7 KA    LIN3  linear TRUE      
#>  8 K     LIN3  linear TRUE      
#>  9 V     LIN3  linear TRUE      
#> 10 KA    RND1  linear TRUE      
#> # … with 29 more rows