The Past, Present, and Future of Macroeconomic Forecasting
نویسنده
چکیده
Broadly defined, macroeconomic forecasting is alive and well. Nonstructural forecasting, which is based largely on reduced-form correlations, has always been well and continues to improve. Structural forecasting, which aligns itself with economic theory and hence rises and falls with theory, receded following the decline of Keynesian theory. In recent years, however, powerful new dynamic stochastic general equilibrium theory has been developed, and structural macroeconomic forecasting is poised for resurgence. Acknowledgments: Helpful input was provided by the panel members and audience at the American Economic Association’s 1997 New Orleans roundtable, “Monetary and Fiscal Policy: The Role for Structural Macroeconomic Models.” I am also grateful to Brad DeLong, Robert Fildes, Lutz Kilian, Alan Krueger, Ben McCallum, Timothy Taylor, Ken Wolpin, Mike Woodford, and seminar participants at the Federal Reserve Bank of San Francisco, especially Tim Cogley, Ken Kasa, Glenn Rudebusch and Tom Sargent. I, however, bear full responsibility for all remaining flaws. The National Science Foundation, the Sloan Foundation and the University of Pennsylvania Research Foundation provided research support. -1The reports of the death of large-scale macroeconomic forecasting models are not exaggerated. But many observers interpret the failure of the early models as indicative of a bleak future for macroeconomic forecasting more generally. Such is not the case. Although the large-scale macroeconomic forecasting models didn’t live up to their original promise, they nevertheless left a useful legacy of lasting contributions from which macroeconomic forecasting will continue to benefit: they spurred the development of powerful identification and estimation theory, computational and simulation techniques, comprehensive machinereadable macroeconomic databases, and much else. Moreover, past failures do not necessarily imply a bleak future: we learn from our mistakes. Just as macroeconomics has benefitted from rethinking since the 1970s, so too will macroeconomic forecasting. Understanding the future of macroeconomic forecasting requires understanding the interplay between measurement and theory, and the corresponding evolution of the nonstructural and structural approaches to forecasting. Nonstructural macroeconomic forecasting methods attempt to exploit the reduced-form correlations in observed macroeconomic time series, with little reliance on economic theory. Structural models, in contrast, view and interpret economic data through the lens of a particular economic theory. Structural econometric forecasting, because it is based on explicit theory, rises and falls with theory, typically with a lag. Structural Keynesian macroeconomic forecasting, based on postulated systems of decision rules, enjoyed a golden age in the 1950s and 1960s, following the advances in Keynesian theory in the 1930s and 1940s, and the two declined together in the 1970s and 1980s. The evolution of nonstructural economic forecasting, in contrast, is less bound to fashions in economic theory; its origins long predate structural -2Keynesian macroeconomic forecasting, and progress continues at a rapid pace. One is naturally led to a number of important questions. What of the impressive advances in economic theory of the 1980s and 1990s? Should we not expect them to be followed by a new wave of structural macroeconomic forecasting, or has nonstructural forecasting permanently replaced structural forecasting? Related, is it necessary to choose between the structural and nonstructural approaches, or might the two be complements rather than substitutes? If a new structural forecasting is likely to emerge, in what ways will it resemble its ancestors? In what ways will it differ? Our answers will take us on a whirlwind tour of the past, present and future of both structural and nonstructural forecasting. We’ll begin by tracing the rise and fall of the structural Keynesian system-of-equations paradigm, and then we’ll step back to assess the long-running and ongoing progress in the nonstructural tradition. Finally, we’ll assess the rise of modern dynamic stochastic general equilibrium macroeconomic theory, its relationship to nonstructural methods, and its implications for a new structural macroeconomic forecasting. 1. The Rise and Fall of Keynesian Macroeconomic Theory and Structural Forecasting Some important forecasting situations involve conditional forecasts; that is, forecasts of one or more variables conditional upon maintained assumptions regarding, for example, the behavior of policy makers. Conditional forecasts require structural models. Structural econometrics, and hence structural macroeconomic forecasting, makes use of macroeconomic theory, which implies that developments in structural forecasting naturally lag developments in theory. The first major wave of twentieth century macroeconomic theory was the Keynesian theory of the 1930s and 1940s, and it was followed by major advances in structural
منابع مشابه
Using Methods Based on Neural Networks to Predict and Manage Diseases (A Case Study of Forecasting the Trend of Corona Disease)
Aim and background: Forecasting methods are used in various fields; one of the most important fields is the field of health systems. This study aimed to use the Artificial Neural Network (ANN) method in forecasting Corona patients in Iran. Method: The present study is descriptive and analytical of a comparative type that uses past information to predict the future, the time series of Corona in...
متن کاملDifferent Methods of Long-Term Electric Load Demand Forecasting a Comprehensive Review
Long-term demand forecasting presents the first step in planning and developing future generation, transmission and distribution facilities. One of the primary tasks of an electric utility accurately predicts load demand requirements at all times, especially for long-term. Based on the outcome of such forecasts, utilities coordinate their resources to meet the forecasted demand using a least-co...
متن کاملIncorporation of Social Media Data into Macroeconomic Forecast Systems: A Mixed Frequency Modelling Approach
Macroeconomic forecasts enable the policy-makers to foresee the future economic trends and take prompt measures to ensure longer economic growth and quicker economic recovery. Accurate and timely macroeconomic forecasts may also help the enterprises to make better long-term business strategies. Social media has a swift and sensitive response to economic dynamics through online news reports, int...
متن کاملThe Effect of Uncertainty of Macroeconomic Indicators on Tehran Stock Exchange Return With an Approach of the TVP-SV Model
One of the most important duties of financial economy is modeling and forecasting the volatilities of price of risky assets. From analysts and policy makers’ view, price volatility is a key variable contributing to perception of market volatilities. Therefore, analysts need to have an appropriate of forecast of price volatility as a necessary input to perform duties such as risk management, por...
متن کاملTeachers’ Professional Competencies: Past, Present, and Future
Teachers’ Professional Competencies: Past, Present, and Future M. Rezaai, Ph.D.* The purpose of this paper is to review the expected professional competencies throughout the history of teacher training in Iran. As such it covers both the past, covering the period from teacher training inception in 1918 to the Islamic revolution in 1979; and the present, since the revolution. Of course t...
متن کامل