PARALLEL PLOT FOR THE SUM OF TWO RANDOM VARIABLES DISTRIBUTION

• Main Author: NUROHMAH (NIM : 10108105); Pembimbing : Khreshna I.A. Syuhada, M.Sc, Ph.D, EVI
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/15581
• Institution: Institut Teknologi Bandung

Parallel Plot (Parallel Coordinate) is a modification of the plot (classic) that can visualize high dimensional geometry and multivariate data analysis. The advantages <br /> <br /> <br /> of using this parallel plot are; connected to many dimension shown in limited space and can see the trends/patterns in the data. Both of these advantages, in particular, will be applied to see the tendency how the pattern of independence and dependence two random variables in parallel plot. Moreover, it will be applied in bivariate distribution and sum of two random variables distribution, that is for two Poisson random variables and two Normal random variables. To develop sum of two random variables distribution, the joint (bivariate) distribution is required. Bivariate Poisson distribution can be constructed <br /> <br /> <br /> through the creation of two new random variables from the sum of two independent Poisson random variables. Bivariate Normal distribution can be constructed through both of conditional distribution and Farlie-Gumbel-Morgenstern copula. From the overall distribution, obtained the characteristics of distribution such as the probability density function (pdf), the cumulative density function (cdf), mean, variance, and identifying feature of the k-th moment that is represented in parallel plot visualization.