Implementation of methods Extremum Surface Estimator (ESE) and
Extremum Distance Estimator (EDE) to identify the inflection point of a curve .
Christopoulos, DT (2014) .
Christopoulos, DT (2016) < https://veltech.edu.in/wp-content/uploads/2016/04/Paper-04-2016.pdf> .
Christopoulos, DT (2016) .
Because it does not imply any kind of functional hypothesis for the data under examination
Because it can give you an estimation despite the level of noise added to initial data
Because it is fast and can use parallel computing if you ask for it
Due to its simplicity it can handle data sets with more than a million rows in negligible execution time
It uses sophisticated iterative methods like bisection of Numerical Analysis and locates the inflection point when it is not directly visible form the first sight
News
What's new in version 1.3
Use of parallel computing under request (doparallel=TRUE) for the functions:
ese()
bese()
findiplist()
findipiterplot()
Repair of function findipiterplot()
Fix bugs in calling the function
Remove indirect methods CRESE, CREDE due to their limited functionality
Plot in two separate pdfs 'ese_iterations.pdf' and 'ede_iterations.pdf'
Check for the existence of sufficient number of results before creating confidence intervals
What's new in version 1.2
The function eixf(x, y, f, i) was removed as not essentially necessary
New functions with self declaring names were added:
*ese(x,y,index)
*bese(x,y,index)
*ede(x,y,index)
*edeci(x,y,index)
*bede(x,y,index)
All functions require length(x)>=4 in order to create numeric output
Thef unction findipiterplot(x,y,index) was improved and became
findipiterplot(x, y, index, plots = TRUE, crossrun = FALSE, ci = FALSE)