Business & Economics
Modelling and Comparing the Forecastability of CPI and PCE Measures of Inflation
Document Type
Oral Presentation
Location
Indianapolis, IN
Subject Area
Business & Economics
Start Date
13-4-2018 11:15 AM
End Date
13-4-2018 11:45 AM
Sponsor
Heather Tierney (Indiana University-Purdue University Fort Wayne)
Description
The Federal Reserve uses Total and Core Personal Consumption Expenditure (PCE) to model inflation. On the other hand, since new outlets tend to report inflation using Total and Core Consumer Price Index (CPI), the general public is more familiar with the measure of inflation using the different forms of CPI. This paper investigates the differences and similarities of modeling and forecasting with Total and Core PCE and Total and Core CPI. Model selection for forecasting is extremely important because it aids in the understanding of the underlying relationship of the data. This is essential for decision making since economic policy is being formed and then applied to real-world events. Model selection also helps us to explain the past relationship between variables and it can also be beneficial for forecasting purposes. This paper uses atheoretical models of annualized quarterly inflation from Total and Core PCE and Total and Core CPI for the time period of 1984:Q1 to 2018:Q4 using the Autoregressive Integrated Moving Average (ARIMA) framework.
Modelling and Comparing the Forecastability of CPI and PCE Measures of Inflation
Indianapolis, IN
The Federal Reserve uses Total and Core Personal Consumption Expenditure (PCE) to model inflation. On the other hand, since new outlets tend to report inflation using Total and Core Consumer Price Index (CPI), the general public is more familiar with the measure of inflation using the different forms of CPI. This paper investigates the differences and similarities of modeling and forecasting with Total and Core PCE and Total and Core CPI. Model selection for forecasting is extremely important because it aids in the understanding of the underlying relationship of the data. This is essential for decision making since economic policy is being formed and then applied to real-world events. Model selection also helps us to explain the past relationship between variables and it can also be beneficial for forecasting purposes. This paper uses atheoretical models of annualized quarterly inflation from Total and Core PCE and Total and Core CPI for the time period of 1984:Q1 to 2018:Q4 using the Autoregressive Integrated Moving Average (ARIMA) framework.