The global economy is growing at the slowest pace in nearly 17 years, the World Bank says. International discord over trade and rising interest rates are among the factors weighing on growth.
Forecasting is a key activity in the field of economics, providing critical inputs to the decision making of central banks, fiscal authorities, and private sector agents. One of the most important outcomes is a forecast of national output (GDP), the measure that includes the goods and services produced in a nation. GDP forecasts are produced by governments and by private entities, such as companies and universities. A wide range of techniques are employed to produce these output forecasts, ranging from judgmental methods that rely on the expertise of individual forecasters to adjust forecasts produced by a suite of models to dynamic stochastic general equilibrium (DSGE) model-based approaches that are disciplined by modern economic theory.
The most commonly used technique for forecasting is time series modeling. This methodology requires a statistical characterization of the behavior of a variable over a period, usually several years, and a belief that the estimated patterns of that variable’s behavior will persist in the future.
The time series approach also requires assumptions about many other variables that can influence an economy over the long term, such as population pressures, the introduction of new products, or changes in financial institutions. These assumptions must be weighed against a knowledge that economic forecasts are heavily influenced by a forecaster’s own theories about how the economy works, which can lead to subjective or biased projections.