num_diff: Function for numeric differentiation

In Chitran1987/StatsChitran: StatsChitran

Function for numeric differentiation

Description

Returns a dataframe consisting of the numerically differentiated second variable w.r.t the first variable to the nth order (d^n Y/d X^n) as the second column and (X) as the first(independent variable) column. It also plots the curve depending on the input arguments

Usage

num_diff(X,Y)
num_diff(X,Y, pl=F)
num_diff(X,Y, type='left', pl=T)
num_diff(X,Y, type='right', pl=T)
num_diff(X,Y, type='center', pl=T)
num_diff(X Y, type='left', order=3)

Arguments

X	The input X-column.
Y	The input Y-column.
type	The type variable can be set to 'left', 'right' or 'center'. Depending on each it calculates the left handed or right handed differentiation dataframe. Set to 'center' by default. Note, when type='center', the X-axis values of the returned dataframe differs slightly from the x-axis values of the original dataset.
pl	Set to TRUE by default, the argument can be toggled.When set to TRUE, Outputs a plot of the differentiated function with respect to the original. The legend indicates the original plot ns the derivative. The no. in the bracket of the legend of the derivative indicates the order of differentiation.
order	Needs to be an integer and a scalar. Fixes the order n of the differentiation. Default value set to 1. That is, calculates the derivative of the first order (dY/dX) when the order argument is left unspecified.

Details

The function always returns a dataframe. There are no other return types or no way to tune the return type in this function.

Value

The return type of the function is always a 2-column dataframe with the 1st column being the independent variable, while the next column is the dependent variable(dY/dX)

Author(s)

Chitran Ghosal

Examples

library('StatsChitran')

## define the base dataset
X <- seq(0,2*pi, by=0.01)
y <- sin(X)

##divide the plot area into two rows
subplot(c(2,1))

##first order derivative plot
a <- num_diff(X,y)
abline(v=pi, col='red')
abline(h=0, col='black')

##second order derivative plot
b <- num_diff(X, y, order = 2)
abline(v=pi, col='red')
abline(h=0, col='black')

##revert back to original plot area
subplot(c(1,1))

Chitran1987/StatsChitran documentation built on Feb. 23, 2025, 8:30 p.m.