The aba control which determines how an aba summary will be calculated and printed to console.
Usage
aba_control(
include_intercept = FALSE,
include_covariates = TRUE,
pval_digits = 4,
aic_digits = 0,
metric_digits = 2,
coef_digits = 2
)Arguments
- include_intercept
boolean. Whether to include intercept in coefs
- include_covariates
boolean. Whether to include covariates in coefs
- pval_digits
integer. How many decimals of a pvalue to show
- aic_digits
integer. How many decimals of AIC value to show
- metric_digits
integer. Default value of how many decimals to show for model metrics (e.g., auc, adj.r.squared, etc)
- coef_digits
integer. Default value of how many decimals to show for model coefficients
Examples
df <- adnimerge %>% dplyr::filter(VISCODE == 'bl')
# standard example
model <- df %>% aba_model() %>%
set_groups(everyone()) %>%
set_outcomes(CSF_ABETA_STATUS_bl) %>%
set_predictors(
PLASMA_PTAU181_bl, PLASMA_NFL_bl,
c(PLASMA_PTAU181_bl, PLASMA_NFL_bl)
) %>%
set_covariates(AGE, GENDER, EDUCATION) %>%
set_stats('glm') %>%
aba_fit()
#> [1] "CSF_ABETA_STATUS_bl ~ AGE + GENDER + EDUCATION"
#> [1] "CSF_ABETA_STATUS_bl ~ AGE + GENDER + EDUCATION + PLASMA_PTAU181_bl"
#> [1] "CSF_ABETA_STATUS_bl ~ AGE + GENDER + EDUCATION + PLASMA_NFL_bl"
#> [1] "CSF_ABETA_STATUS_bl ~ AGE + GENDER + EDUCATION + PLASMA_PTAU181_bl + PLASMA_NFL_bl"
# no control -> default
model_summary <- model %>% aba_summary()
print(model_summary)
#> ---------------------------------------------------------
#> Group: Everyone | Outcome: CSF_ABETA_STATUS_bl | Stat: S1
#> ---------------------------------------------------------
#> Coefficients & Metrics:
#> # A tibble: 4 × 10
#> predictor AGE GENDER EDUCATION PLASMA_PTAU181_bl PLASMA_NFL_bl auc
#> <chr> <chr> <chr> <chr> <chr> <chr> <chr>
#> 1 Basic 1.03 [… 1.39 … 0.93 [0.… NA NA 0.58…
#> 2 M1 1.01 [… 1.31 … 0.96 [0.… 3.89 [2.80, 5.52… NA 0.72…
#> 3 M2 1.00 [… 1.43 … 0.94 [0.… NA 2.27 [1.48, … 0.63…
#> 4 M3 1.00 [… 1.33 … 0.96 [0.… 3.67 [2.61, 5.29… 1.29 [0.81, … 0.72…
#> # ℹ 3 more variables: aic <chr>, pval <chr>, nobs <chr>
#>
# add a control object - don't include covariate coefficients
my_control <- aba_control(include_covariates = FALSE)
model_summary2 <- model %>% aba_summary(control = my_control)
print(model_summary2)
#> ---------------------------------------------------------
#> Group: Everyone | Outcome: CSF_ABETA_STATUS_bl | Stat: S1
#> ---------------------------------------------------------
#> Coefficients & Metrics:
#> # A tibble: 3 × 7
#> predictor PLASMA_PTAU181_bl PLASMA_NFL_bl auc aic pval nobs
#> <chr> <chr> <chr> <chr> <chr> <chr> <chr>
#> 1 M1 3.89 [2.80, 5.52] (P<0.… NA 0.72… 799 <0.0… 645
#> 2 M2 NA 2.27 [1.48, … 0.63… 862 0.00… 645
#> 3 M3 3.67 [2.61, 5.29] (P<0.… 1.29 [0.81, … 0.72… 800 <0.0… 645
#>