Time Series Modeling and Forecasting of US Consumer Price Index Using WTI Crude Oil Price

Abstract

This study aims to forecast U.S. inflation by using crude oil prices as a key variable. WTI crude oil price and Consumer Price Index Year-over-Year (CPI YoY) data from 2000 to 2022 are extracted to construct a time series model. The empirical analysis relies on Autoregressive Integrated Moving Average (ARIMA) model and regression to formulate forecasts based on historical patterns and error dynamics. The study finds out that compared to the automatically generated ARIMA model, using fitted values of the WTI time series model to predict CPI YoY through a multi-variable model is better. According to the prediction, the CPI YoY forecast for 2023 reveals a decreasing tendency in 2023, implying a slower rate of increase in U.S. inflation. The result provides insights into economic conditions, helps decision-making, and mitigates the potential risks associated with price fluctuations. Overall, the research contributes to a deeper understanding of the dynamics between energy markets and inflationary pressures in the United States.