Forecasting, often found alongside budgeting and planning processes, uses past and present data, trend analysis, and executive insight in order to predict the future state of any given metric.
There are three different techniques in forecasting:
- Qualitative: market research or the delphi method (collection of expert opinions), panel census
- Time series analysis and projection: historical data, trend projections Box Jenkins, X-11
- Causal models: regression model, econometric model, input-output model, life cycle analysis
Like the weather, financial forecasting isn’t an exact science and indicating uncertainty in forecasts is common practice. In order to mitigate errors, access to accurate historical and real-time data greatly improves the accuracy of forecasts. In addition, the ability to combine rolling forecasts, long-range forecasts, scenario playing and stress test events helps bolster forecast findings and makes forecasts more agile and responsive to business, economic or KPI change.