Version 0.5.0

Added function for computing “one-sample” equivalency factors. This function should produce the same values as cmstatr::k_equiv(), but is written in C++ and is hence faster.
Renamed the function p_equiv() to p_equiv_one_sample(). The function p_equiv() will still work but will produce a warning and may be removed in a future version of this package.
Reduced the tolerance for root finding from DBL_EPSILON^0.25 (about 1.2e-4) DBL_EPSILON^0.5 (about 1.5e-8). This may affect the results of k_equiv_two_sample, p_equiv_two_sample and p_equiv_one_sample and should produce more accurate results for each of these functions. You can expect to see the largest differences with very low p-values and alpha values.
In the function power_sim_dual(), changed the internal data types used to count the number of acceptance/equivalency failures from int to unsigned long long to prevent overflow when replicates is greater than 46340.
This package now checks for conflicting function names and will produce a message when attached, similar to the way that the tidyverse package produces a conflict message when attached.
In the accompanying website [https://cmstatrExt.cmstatr.net], added a new calculator for equivalency thresholds.

Version 0.4.1

Minor update to call Rcpp::stop in lieu of Rf_error due to an upcoming change to the Rcpp package. No changed in functionality.

Added functions for creating average stress-strain curves (average_curve_lm and average_curve_optim)
Added example stress-strain data (pa12_tension and fff_shear)
Added vignette with examples of fitting average curves to stress-strain data