Time Series Methodologies towards Finanical Data Anaysis
Open Access
Author:
Jiang, Yunjie
Area of Honors:
Statistics
Degree:
Bachelor of Science
Document Type:
Thesis
Thesis Supervisors:
Dr. Zhibiao Zhao, Thesis Supervisor David Russell Hunter, Thesis Honors Advisor
Keywords:
Time Series MA model AR model ARIMA model ARCH/GARCH model Volatility
Abstract:
The question of predicting future value based on historical data always motivates econometricians and investors in both academia and real industry. The volatility of financial data challenges thousands of investors from making the right decision while investing. Finding the right techniques to solve the problems is critical to investors and scholars.
Time series analysis, therefore, has been developed to diagnose the correlated relationships among variables. In this thesis, I will introduce some techniques of time series from a statistical perspective including MA model, AR model, ARIMA model, and ARCH/GARCH models that are widely used in econometrics and financial world. These statistical methods help us to understand the relationships between previous and future data.