Forecasting US Inflation Trends: Insights from Time Series Analysis

Abstract

In our research, we dive into the world of forecasting US inflation using the ARIMA (Auto-Regressive Integrated Moving Average) model. We crafted a predictive framework by meticulously analyzing data, conducting tests like the Augmented Dickey-Fuller, and crunching metrics such as RMSE and MSE. Our meticulous model selection procedures involve carefully examining many ARIMA configurations to identify the ideal parameters, ensuring robustness and accuracy in forecasting inflation trends over the study period. Our findings show that the ARIMA (0,1,2) model outperforms others, offering reliable forecasts. These results not only showcase the model\'s effectiveness but also provide valuable insights for both policymakers and market players. By understanding potential inflationary trends, they can confidently navigate risks and make informed decisions, ultimately fostering stability and growth in the economy.

Introduction

I. INTRODUCTION

Inflation is a fundamental economic phenomenon that profoundly influences various aspects of the economy, ranging from consumption patterns to investment decisions and monetary policy formulation. Understanding the dynamics of inflation and its underlying drivers is crucial for policymakers, economists, and market participants alike. This paper presents a comprehensive time series analysis of inflation in the United States spanning the period from 1947 to 2022.

Over the past decades, the US economy has experienced significant fluctuations in inflation rates, ranging from periods of high inflationary pressures to periods of subdued inflation or even deflationary concerns. These fluctuations have been influenced by a multitude of factors, including changes in monetary policy stance, fiscal policy measures, oil price shocks, and broader macroeconomic trends.

The study of inflation dynamics is inherently complex due to its multifaceted nature and the interplay of various economic forces. Time series analysis offers a powerful framework for examining the behavior of inflation over time, allowing researchers to identify long-term trends, cyclical patterns, and potential drivers.

The primary objectives of this study are twofold: first, to provide a comprehensive understanding of inflation dynamics in the US over the specified historical period, and second, to assess the forecasting performance of the developed models in predicting future inflationary trends. By achieving these objectives, the main aim of this research is to contribute valuable insights to the current literature on inflation analysis and provide practical guidance for policymakers and market participants in managing inflation risks and making informed decisions.

Financial time series can be forecasted using a variety of techniques. Univariate forecasting is one technique that solely estimates time. The autoregressive integrated moving average, or ARIMA model is a unique kind of modeling where the moving average component (the moving average component) of the time series is determined by taking the historical values of the autoregressive component and adding the current and lag values of the white noise error term. The ARIMA model and how we may forecast or anticipate future values based on past values are the main topics of the paper. In the subsequent sections of this paper, we will delve into the methodology employed for data collection and analysis, present the findings of our time series analysis, discuss the implications of our results, and conclude with recommendations for future research and policy considerations.

II. LITERATURE SURVEY

The study of inflation dynamics has been the subject of extensive research in economics literature, with scholars employing various methodologies to understand the complexities of inflation behavior. Numerous studies have investigated inflation trends, drivers, and forecasting techniques, offering valuable insights into the dynamics of inflation in different economic contexts.

One seminal work in this field is the study conducted by Fisher (1972), which emphasized the role of monetary factors in driving inflationary pressures.

Fisher's analysis highlighted the significance of money supply growth as a key determinant of inflation, laying the groundwork for subsequent research on the monetary theory of inflation. However, while Fisher's study provided valuable insights into the relationship between money supply and inflation, it has been criticized for its narrow focus on monetary factors and its neglect of other important determinants of inflation, such as fiscal policy measures and supply-side shocks (Taylor, 1979). Taylor argued that a more comprehensive approach to inflation analysis is needed to capture the full range of factors influencing inflation dynamics.

In response to these criticisms, subsequent research has expanded the scope of inflation analysis to include a broader set of variables and factors. For example, Blinder (1997) conducted a comprehensive study of inflation dynamics in the United States, examining the impact of various macroeconomic variables, including monetary policy, fiscal policy, oil price shocks, and exchange rate movements.

Blinder's study provided valuable insights into the multifaceted nature of inflation dynamics and underscored the importance of considering multiple factors in inflation analysis. However, one of the drawbacks of Blinder's study was its reliance on traditional econometric models, which may have limitations in capturing the nonlinear and dynamic nature of inflation processes (Stock & Watson, 2007).

Recent developments in time series analytic methods have created new avenues for a more thorough examination of inflation dynamics. Researchers have increasingly turned to sophisticated econometric models, such as ARIMA, to analyze inflation time series data and uncover hidden patterns and trends.

Overall, the literature on inflation analysis provides a rich body of knowledge that continues to evolve with advancements in economic theory and empirical techniques. While past studies have contributed valuable insights into inflation dynamics, there remains a need for further research to refine existing models, incorporate new data sources, and improve forecasting accuracy.

III. MATERIALS AND METHOD

George Box and Gwilym Jenkins created the ARIMA model in the 1970s and used mathematics to describe changes in time series data. The terms ARIMA and Box-Jenkins are used interchangeably in certain instances. A statistical analysis technique called Auto-regressive Integrated Moving Average, or ARIMA, makes use of time series data to forecast future trends or to get a deeper understanding of the data set. If a statistical model forecasts future values by using historical data, it is said to be autoregressive. An ARIMA model could, for instance, attempt to estimate a company's profitability based on historical periods or predict a stock's future pricing based on its historical performance. Lagged moving averages are used by ARIMA to smooth time series data.

Auto-Regressive (AR): The abbreviation for autoregressive models is AR. The time-series' lagged values determine the future values. The AR model estimates future data values by looking at historical data values. The order "p" indicates that current values are predicted using previous data from period "p." The following is a pth-order AR process:

VI. ACKNOWLEDGEMENTS

We express sincere gratitude to all contributors to this research on "FORECASTING US INFLATION TRENDS: INSIGHTS FROM TIME SERIES ANALYSIS”. Special thanks to Ms. Raheel Hassan for invaluable guidance and mentorship. We thank Chandigarh University for resources and this opportunity to work on this. We would like to acknowledge the contributions of the researchers whose seminal works and studies lays the foundation for our research. Their insights and findings have enriched our understanding of inflation dynamics and guided our analytical approach. We appreciate Kaggle.com for data access. Lastly, thanks to friends, colleagues, and family for unwavering support.

Conclusion

In this paper, we analyzed US inflation dynamics from 1947 to 2022, using advanced time series techniques. We tested data stationarity with the Augmented Dickey-Fuller (ADF) test, then applied differencing for stationarity. Next, we employed ARIMA for forecasting, testing various models to find the best fit based on AIC, BIC, RMSE, and MSE. Having selected the most suitable ARIMA model, we proceeded to generate inflation forecasts for the specified period. Visualizations, including line plots and residual plots, were utilized to present the forecasted results and assess the model\'s performance. In conclusion, our research offers a strong forecasting framework for inflation trends, benefiting policymakers, economists, and market participants. However, it is essential to acknowledge the limitations of our study, including the assumptions underlying the ARIMA model and potential uncertainties in forecasting future inflation trends. Future research may explore alternative modeling approaches, incorporate additional variables, or employ more sophisticated techniques to enhance forecasting accuracy.

References

[1] Fisher, I. (1972). The Theory of Interest. Dover Publications. [2] Taylor, J. B. (1979). Staggered wage setting in a macro model. The American Economic Review, 69(2), 108-113. [3] Blinder, A. S. (1997). Monetary policy and the inflation process: Some lessons from the 1980s. Economic Review, Federal Reserve Bank of Kansas City, 82(4), 5-22. [4] Stock, J. H., & Watson, M. W. (2007). Introduction to Econometrics (3rd ed.). Addison Wesley. [5] Box, George E. P. Time series analysis: forecasting and control. -- Fifth edition / George E.P. Box, Gwilym M. Jenkins, Gregory C. Reinsel, Greta M. Ljung. [6] Pincheira P, Medel CA (2012b): Forecasting Inflation with a Simple and Accurate Benchmark: A Cross-Country Analysis. Central Bank of Chile, Working Paper, no. 677. [7] Proietti T (2011): Direct and Iterated Multistep AR Methods for Difference Stationary Processes. International Journal of Forecasting, 27(2):266-280. [8] Stock JIH, Watson MW (1996): Evidence on Structural Stability in Macroeconomic Time Series Relations. Journal of Business and Economic Statistics, 14:11-30. [9] Wagenmakers E-J, Farrell S (2004): AIC Model Selection Using Akaike Weights. Psychonomic Bulletin and Review, 11:192-196. [10] West K (1996): Asymptotic Inference about Predictive Ability. Econometrica, 64:1067-1084. [11] Cosgrove, J. (2020). The Russian invasion of the Crimean peninsula, 2014-2015: A post-cold war nuclear crisis case study. Applied Physics Lab, Johns Hopkins University Laurel Md Laurel. [12] Ara Stepanyan, Agustin Roitman, Gohar Minasyan, Dragana Ostojic, and Natan Epstein.(2015). Spillover. THE SPILLOVER EFFECTS of Russia’s Economic Slowdown on Neighboring Countries. [13] Russia inflation rate in 2014. Russian Federation Inflation Rate in 2014. (n.d.). [14] Ozili, P. K. (2022). Global economic consequence of Russian invasion of Ukraine.SSRN, 4064770. [15] Yurina, E. A., Tishchenko, E. S., Fedotova, E. V., Markov, B. E., & Kuzmenko, N. I. (2022). New Approaches to the Economic Security of Russian Regions in Terms of the COVID-19 Pandemic. In Geo-Economy of the Future: Sustainable Agriculture and Alternative Energy. Cham: Springer International Publishing, 699-715. [16] Kasyan, E. A., & Nemirovich-Danchenko, M. M. (2022). Selection of a forecasting model for Russia consumer price index based on the application of Analytic Hierarchy Process. In Journal of Physics: Conference Series, 2291, 1, 012006. [17] Benkraiem, R., Garfatta, R., Lakhal, F., & Zorgati, I. (2022). Financial contagion intensity during the COVID-19 outbreak: A copula approach. International Review of Financial Analysis, 102136. [18] GAUTAM, R. S., & KANOUJIYA, J. (2022). Inflation Targeting: An Application of ARIMA Modelling Using Forecasting of CPI and WPI. Iconic Research and Engineering Journals, 5(11), 195-198. [19] Jalil, A., & Rao, N. H. (2019). Time series analysis (stationarity, cointegration, and causality). In Environmental Kuznets Curve (EKC).Academic Press, 85-99.

Copyright

Copyright © 2024 Raheel Hassan, Kanishk Bhandari, Rajkumar Singh, Yokshit Sankhyan, Deepanshu Kumar, Yash Gupta. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.