Value at Risk Estimation with Hybrid-SVR-GARCH-KDE Model for LQ45 Portfolio Optimization

Shindi Shella May Wara*, Dedy Dwi Prastyo, Heri Kuswanto

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

5 Citations (Scopus)

Abstract

Stock is a financial instrument that has a high variation. One way to determine the risk of a stock is to estimate the Value at Risk. However, Value at Risk cases tend to have fluctuating variations over time and are difficult to model because they are hypothesized to be non-linear. To capture this, modeling is carried out with Generalized Autoregressive Conditional Heteroscedasticity (GARCH). Meanwhile, identification of non-linear models can be solved using machine learning methods, one of which is Support Vector Regression (SVR) which is sensitive to over fitting cases. To produce an optimal model, reinforced with Kernel Density Estimation (KDE). By using this combination, the hybrid SVR-GARCH-KDE method is obtained. The results of this method can show that the Hybrid-SVR-GARCH-KDE method is good for estimating the Value at Risk on return of LQ45 stock price data for the period January 2018 to March 2021 which has the smallest PEB and PBV. From this method, portfolio optimization was carried out and resulted in a decision that investments in the Trade, Services, and Investment sectors as well as mining were profitable for investors.

Original languageEnglish
Title of host publication3rd International Conference on Science, Mathematics, Environment, and Education
Subtitle of host publicationFlexibility in Research and Innovation on Science, Mathematics, Environment, and Education for Sustainable Development
EditorsNurma Yunita Indriyanti, Meida Wulan Sari
PublisherAmerican Institute of Physics Inc.
ISBN (Electronic)9780735443099
DOIs
Publication statusPublished - 27 Jan 2023
Event3rd International Conference on Science, Mathematics, Environment, and Education: Flexibility in Research and Innovation on Science, Mathematics, Environment, and Education for Sustainable Development, ICoSMEE 2021 - Surakarta, Indonesia
Duration: 27 Jul 202128 Jul 2021

Publication series

NameAIP Conference Proceedings
Volume2540
ISSN (Print)0094-243X
ISSN (Electronic)1551-7616

Conference

Conference3rd International Conference on Science, Mathematics, Environment, and Education: Flexibility in Research and Innovation on Science, Mathematics, Environment, and Education for Sustainable Development, ICoSMEE 2021
Country/TerritoryIndonesia
CitySurakarta
Period27/07/2128/07/21

Fingerprint

Dive into the research topics of 'Value at Risk Estimation with Hybrid-SVR-GARCH-KDE Model for LQ45 Portfolio Optimization'. Together they form a unique fingerprint.

Cite this