ANALYSIS OF KERNEL-BASED COPULA PROCESSES MODEL AND ITS APPLICATIONS
Prediction of time series data is an effort to determine the value of the data in the future by using historical data. The results of this prediction is very useful both in financial data or non - financial data. Distribution function prediction can be obtained using Bayesian concept of joint dis...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Subjects: | |
| Online Access: | https://digilib.itb.ac.id/gdl/view/33880 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Prediction of time series data is an effort to determine the value of the data
in the future by using historical data. The results of this prediction is very useful
both in financial data or non - financial data. Distribution function prediction can
be obtained using Bayesian concept of joint distribution function. Construction
of joint distribution function from some specific distributed random variables is
certainly not easy to do. So we propose to use Copula function to build the joint
distribution function. Kernel - based Copula Processes (KCP) is a process that uses
the kernel as component parameter of copula Elliptical. This kernel acts as a value
of the correlation or dependence of data that has certain characteristics. This thesis
studied further about KCP models to determine the distribution of predictions using
univariate and multivariate analysis. The distribution of predictions can be used to
predict the time series data for the next period of time. |
|---|