Predictability of Treasury Yield via ARMA and VAR Approach —-Evidence from United States
Open Access
Author:
Zheng, Huguang
Area of Honors:
Economics
Degree:
Bachelor of Science
Document Type:
Thesis
Thesis Supervisors:
Jenny Xiaoe Li, Thesis Supervisor Dr. Russell Paul Chuderewicz, Thesis Honors Advisor
Keywords:
Yield Rate ARMA VAR Predictability
Abstract:
Yield rate of sovereign bonds, as an important indicator of a country’s business cycle, and cost and ability of local governments to raise funds, has also been a policy target for governments and central banks for long. My long-term interest in financial market and monetary policy inspired me to think further about the factors determining the yield rates of sovereign bonds and the predictabil- ity of yield rate based on historical data. This paper will focus on the yield rates of 3-month, 5-year and 10-year U.S treasuries. It aims to shed a light on how different economic variables would affect the course of short-term, mid-term and long-term treasury yield rates. Based on the analysis of the interaction between yield rates and those economics variables, it also makes the attempts to predict the forementioned yield rates by constructing ARMA (auto-regressive and moving average) and VAR (vector auto-regressive) model, and reach a conclusion on its predictability.