Rounding with number of digits determined by an equivalence test

Description

This function implements the suggestions of Astier & Wolak for the number of decimal digits to keep for coefficient estimates. The other statistics are rounded by fmt_significant().

Usage

fmt_equivalence(conf_level = 0.95, digits = 3, pdigits = NULL, ...)

Arguments

conf_level	Confidence level to use for the equivalence test (1 - alpha).
digits	Number of significant digits to keep.
pdigits	Number of decimal digits to keep for p values. If NULL, the value of digits is used.
…	Additional arguments are passed to the format() function (e.g., big.marks, scientific). See ?format

References

Astier, Nicolas, and Frank A. Wolak. Credible Numbers: A Procedure for Reporting Statistical Precision in Parameter Estimates. No. w32124. National Bureau of Economic Research, 2024.

Examples

library("modelsummary")

library(modelsummary)
mod <- lm(mpg ~ hp, mtcars)

# Default equivalence-based formatting
modelsummary(mod, fmt = fmt_equivalence())

	(1)
(Intercept)	30
	(1.63)
hp	-0.07
	(0.0101)
Num.Obs.	32
R2	0.602
R2 Adj.	0.589
AIC	181.2
BIC	185.6
Log.Lik.	-87.619
F	45.460
RMSE	3.74

# alpha = 0.2
modelsummary(mod, fmt = fmt_equivalence(conf_level = .8))

	(1)
(Intercept)	30
	(1.63)
hp	-0.068
	(0.0101)
Num.Obs.	32
R2	0.602
R2 Adj.	0.589
AIC	181.2
BIC	185.6
Log.Lik.	-87.619
F	45.460
RMSE	3.74

# default equivalence, but with alternative significant digits for other statistics
modelsummary(mod, fmt = fmt_equivalence(digits = 5))

	(1)
(Intercept)	30
	(1.6339)
hp	-0.07
	(0.010119)
Num.Obs.	32
R2	0.602
R2 Adj.	0.589
AIC	181.2
BIC	185.6
Log.Lik.	-87.619
F	45.460
RMSE	3.74