Comparing methods of parameter estimation in INGARCH(1,1) model

Hasbullah, Muhammad Hasri and Mohamad, Nurul Najihah (2023) Comparing methods of parameter estimation in INGARCH(1,1) model. In: FYP Exhibition 2022/2023, 2022, IIUM Kuantan.

PDF - Published Version Restricted to Repository staff only Download (1MB) | Request a copy

Abstract

In recent years, integer-valued time series models have gained considerable attention among researchers due to their practical significance. Among various parameter estimation techniques, Maximum Likelihood Estimation (MLE) is widely used. However, studies have highlighted limitations of MLE, particularly regarding consistency and optimality. These concerns have prompted statisticians and econometricians to explore alternative estimation methods. This study investigates three approaches: the classical Least Squares Estimation (LSE), the novel Quadratic Estimating Function (QEF), and MLE. A comparative analysis of these methods is performed using the INGARCH(1,1) model through simulation, evaluating their performance based on Mean Square Error (MSE), Standard Error (SE), and bias. The findings indicate that QEF produces lower SE values compared to both MLE and LSE. For practical application, the INGARCH(1,1) model is fitted to daily COVID-19 case data in Malaysia from 18 March 2020 to 18 March 2022, demonstrating the effectiveness of the QEF method.

Item Type:	Proceeding Paper (Poster)
Uncontrolled Keywords:	MLE, LSE, QEF, integer-valued, INGARCH
Subjects:	Q Science > QA Mathematics > QA276 Mathematical Statistics
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button):	Kulliyyah of Science > Department of Computational and Theoretical Sciences Kulliyyah of Science
Depositing User:	Dr Nurul Najihah Mohamad
Date Deposited:	22 Aug 2025 12:16
Last Modified:	22 Aug 2025 12:16
URI:	http://irep.iium.edu.my/id/eprint/122795

Actions (login required)

View Item