This vignette provides an overview of the “NTLKwIEx” package. The package NTLKwIEx offers a powerful range of statistical tools for analysis,simulation, and computation based on the Topp-Leone Kumaraswamy Inverse Exponential (NTLKwIEx) distribution. This distribution, which combines the properties of the Topp-Leone, Kumaraswamy, a new proposal and inverse exponential distributions,is particularly useful for modeling complex, heterogeneous data present in many scientific disciplines. With the “NTLKwIEx” package, users can estimate the parameters of the NTLKwIEx distribution from datasets, generate random samples according to this distribution, plot histograms and density functions, and calculate specific quantiles.
library(stats)
library(dplyr)
#> Warning: le package 'dplyr' a été compilé avec la version R 4.2.3
#>
#> Attachement du package : 'dplyr'
#> Les objets suivants sont masqués depuis 'package:stats':
#>
#> filter, lag
#> Les objets suivants sont masqués depuis 'package:base':
#>
#> intersect, setdiff, setequal, union
library(ggplot2)
library(NTLKwIEx)
The distribution is particularly useful for modeling data with heavy tails, skewness, and positive values. It is a versatile distribution that can handle diverse characteristics in the data.
The probability density function (PDF) for the NTLKwIEx distribution is given by the formula:
\[ f(x) = 2abm\dfrac{\theta}{x^{2}}\left( -\dfrac{\theta}{x}log\left( \alpha \right)+ exp\left( \dfrac{\theta}{x}\right)\right) \alpha^{exp\left(-\dfrac{\theta}{x}\right) }exp\left( -\dfrac{\theta}{x}\left(1+a\alpha^{exp\left(-\dfrac{\theta}{x}\right)}\right)\right) \left(1-exp\left( -a\dfrac{\theta}{x}\alpha^{exp\left(-\dfrac{\theta}{x}\right)} \right)\right)^{2b-1}\left( 1-\left( 1-exp\left( -\dfrac{a\theta}{x} \alpha^{exp\left(-\dfrac{\theta}{x}\right) }\right)\right)^{2b} \right)^{m-1} \] where \[\left( \theta, \alpha , a , b, m\right) > 0 \]
Let’s calculate the PDF for \(x=1\) , \(\theta=5\), \(\alpha=4\) , \(a=3\) , \(b=2\) and \(m=1\)
The cumulative density function (CDF) for the NTLKwIEx distribution is given by the formula:
\[ F(x)=\left(1-\left( 1-exp\left( -\dfrac{a\theta}{x} \alpha^{exp\left(-\dfrac{\theta}{x}\right) }\right)\right)^{2b} \right)^{m}\] where \[\left( \theta, \alpha , a , b, m\right) > 0 \]
Let’s calculate the CDF for \(x=1\) , \(\theta=5\), \(\alpha=4\) , \(a=3\) , \(b=2\) and \(m=1\)
This function generates a graphical plot of the probability density function (PDF) for the NTLKwIEx distribution.
This function generates a graphical plot of the cumulative density function (CDF) for the NTLKwIEx distribution.
The quantile function calculates the quantile value for a given probability using the NTLKwIEx distribution.
Let’s calculate the quantile using parameters \(p=0.3\) , \(\theta=1.8\), \(\alpha=0.5\) , \(a=5.4\) , \(b=2.4\) and \(m=8.2\)
This function generates random samples from the NTLKwIEx distribution
using the function sapply
.
Let’s generate 100 random samples with parameters \(\theta=5.3\), \(\alpha=3.2\) , \(a=8.2\) , \(b=1.8\) and \(m=2.3\)
set.seed(100)
data=R_NTLKwIEx(100, teta = 5.3, alpha = 3.2, a=8.2, b=1.8, m=2.3)
data
#> [1] 86.82640 80.53718 121.75249 51.41353 108.46004 110.73220 190.76535
#> [8] 94.85908 120.76828 69.34952 135.36246 231.96326 83.37939 98.60411
#> [15] 171.28838 145.00940 73.83060 93.18853 93.44205 150.18030 118.96422
#> [22] 155.55415 119.38661 166.80422 101.55263 69.50365 173.98327 231.79593
#> [29] 121.20020 83.04967 111.41997 281.34138 92.04493 330.66918 151.44860
#> [36] 237.95324 70.69140 136.26983 543.33710 63.83008 89.73133 219.56888
#> [43] 176.61393 197.80977 131.04323 111.86643 177.64258 233.61045 74.22785
#> [50] 86.74053 89.71466 73.06791 77.77668 82.69410 128.75387 80.00035
#> [57] 62.84032 77.04666 129.93809 74.69984 107.74751 140.04541 348.71764
#> [64] 146.76042 105.06677 93.22087 106.58725 105.10470 78.95887 151.21186
#> [71] 100.47143 89.35715 125.30829 370.94944 143.33098 135.30117 214.10252
#> [78] 175.58877 201.23110 57.87579 107.13791 130.28675 269.35716 462.14739
#> [85] 47.10868 126.27966 161.96784 79.41717 85.94000 162.01337 254.57297
#> [92] 74.49665 93.26820 105.51713 254.01657 97.38995 115.97136 63.09704
#> [99] 44.98030 174.51844
Let’s generate 150 random samples with parameters \(\theta=3.1\), \(\alpha=0.2\) , \(a=4.2\) , \(b=2.8\) and \(m=1.3\)
set.seed(100)
data=R_NTLKwIEx(150, teta = 3.1, alpha = 0.2, a=4.2, b=2.8, m=1.3)
data
#> [1] 3.015766 2.891872 3.630332 2.238402 3.409102 3.447897 4.614222 3.167266
#> [9] 3.614404 2.658206 3.843986 5.113648 2.948473 3.235615 4.358706 3.988596
#> [17] 2.754000 3.136357 3.141071 4.064039 3.585028 4.141013 3.591927 4.297769
#> [25] 3.288452 2.661550 4.394928 5.111716 3.621402 2.941960 3.459556 5.656023
#> [33] 3.115000 6.152258 4.082335 5.182361 2.687227 3.857821 7.966051 2.535661
#> [41] 3.071388 4.968544 4.430007 4.703286 3.777448 3.467127 4.443650 5.132632
#> [49] 2.762357 3.014106 3.071071 2.737924 2.835907 2.934921 3.741751 2.881055
#> [57] 2.513101 2.820909 3.760227 2.772239 3.396847 3.914815 6.324628 4.014299
#> [65] 3.350379 3.136959 3.376791 3.351029 2.859991 4.078926 3.269168 3.064290
#> [73] 3.687322 6.531100 3.963804 3.843050 4.903222 4.416370 4.745953 2.397150
#> [81] 3.386316 3.765669 5.529126 7.321931 2.127055 3.702743 4.231079 2.869283
#> [89] 2.998598 4.231712 5.368562 2.767986 3.137839 3.358208 5.362429 3.213637
#> [97] 3.535757 2.518979 2.070163 4.402087 3.063541 3.213734 2.148818 3.145716
#> [105] 3.682792 4.044454 6.688133 4.107274 1.865733 3.584271 4.762542 4.582864
#> [113] 2.353932 2.844162 6.495891 2.126168 5.484980 4.202028 2.743630 4.786651
#> [121] 3.231142 3.221543 3.419269 3.719307 3.123930 2.056439 9.020607 6.240103
#> [129] 3.628471 2.437154 2.841607 4.929315 4.251764 3.482863 3.708471 1.931819
#> [137] 3.418184 2.157442 3.395468 3.859732 4.003269 2.379242 2.578127 5.386472
#> [145] 2.509914 4.212755 6.062540 2.163532 1.971913 2.738996
This function estimates the parameters of the NTLKwIEx distribution while respecting constraints on the parameters.
Let’s estimate parameters from a sample data vector.
Let’s estimate parameters from a sample data vector.
This function estimates parameters and plots the histogram of the data along with the estimated density curve.