I am going to outline forecast following 11 fundamental questions:
- What is forecast?
A forecast is any statement about the future. It can be accurate or inaccurate. It can be made for a 3 months period, an year, 3 years, 5 years, 10 years or longer. Fiscal authorities are increasingly looking at horizons of at least three years.
- What can be forecast?
For monetary policy decisions, for example, inflation (including productivity) and output growth, aggregate demand, future wage developments, and the global outlook.
- How confident can we be in forecasts?
“Because of the future we don’t know that we don’t know, the future is largely unpredictable. But some developments can be anticipated, or at least imagined, on the basis of existing knowledge.” – Maxine Singer (1997).
- How is forecast done in generally?
There are many ways of making forecast. These include formal model based statistical analysis, statistical analysis not based on parametric models, informal “back of the envelope calculations”, simple extrapolations, leading indicators, guessing and “hunches”.
- How is forecast done by economists?
In economics, methods of forecasting include guessing, “rules of thumb”, “informal models”, expert judgment, extrapolation, leading indicators, surveys, time-series models, and econometric system.
- How can one measure the success or failure of forecasts?
A forecast might reasonably be judged “successful” if it was close to the outcome, but that judgment depends on how “close” is measured (accurate and precise).
- How does one analyse the properties of forecast methods?
The properties of forecasting methods can be investigated in both empirical and artificial settings, using mathematical analysis and computer-intensive numerical methods. Series can be compared and often series are selected which shares certain characteristics.
- What special data distinct attributes matter most?
Seasonality, Business-cycle fluctuations, trend growth, successive dependence, and changing variability.
- What are the main problems with forecasting?
Economies evolve over time and are subject to intermittent, and sometimes large, unanticipated shocks. This becomes embodied in physical and human capital, and provides the engine for sustained growth in real output. In addition, structural breaks may be precipitated by changes in legislation, sudden switches in economic policy, or political turmoil.
- Do these problems have potential solutions?
Managing intervals diminish uncertainty.
- What is the future of economic forecast?
To succeed simultaneously in forecasting competitions and in the policy arena, econometric models will have to mimic the adaptability of the best forecasting devices, while retaining their foundations in economic analysis.