Forecasting
Forecasting is defined as the process of predicting the future based on past and current knowledge. Of course, it is impossible to make perfect predictions about the future, but it is believed that it is possible to identify patterns and characteristics in past data that can successfully be adapted to forecast future values (Petropoulos et al. 2022, pp. 710-711). Thus, forecasting in an economical sense is based on the idea of Fama’s Efficient Market Hypothesis (EMH), that prediction of future (financial) values can be made by using past information (Fama 1970, pp. 383-417). Forecasting in a business context can be done regarding to many different objects, such as sales, production, cash flow, and exchange rates. Therefore, forecasting is nowadays a very important tool to cover vital aspects of business and its accuracy also has a serious impact on a company’s activities as well as decision making. Besides the internal impact on the company, such a forecast can also have an influence on the external environment, such as on stock valuation and decreases in customer satisfaction (Klimberg et al. 2010, pp. 137-138). It becomes apparent that forecasting fulfills multiple purposes in a company’s business activities, from its importance for effective and efficient planning up to requirements for capital investment (Hyndman, Athanasopoulos 2021).
Factors that influence forecasting
The accuracy and quality of a forecast highly depend on its subject, as some events are easier to predict than other ones. For instance, the timing of tomorrow’s sunrise can be predicted quite reliably, while the lotto numbers for next week are nearly impossible to forecast. This predictability of a subject is influenced by the following factors:
- Comprehension of a subject’s contribution factors,
- Availability of data,
- Similarity between past and future,
- Possibility to influence the subject that is forecasted.
For example, in case of forecasting currency exchange rates, there is evidently an abundance of data, but it is hard to deduce from the past as there is a high political or economical influence that is hard to assess. And in the case of currencies, it is even more difficult as they are their forecast would be self-fulfilling, because if there was a reliable forecast that predicts its ap - or depreciation, then the market will immediately adjust to close that forecasted change, following the before mentioned EMH. This example shows the importance of knowing whether it is possible to forecast a subject accurately by identifying patterns and dependencies, or whether it is just a coin flip (Hyndman, Athanasopoulos 2021).
Rolling forecast as a new way of budgeting
An especially interesting application of the concept of forecasting can be found in the so-called ‘rolling forecast’, which is a nowadays commonly used tool in business practice. The rolling forecast represents a new way of budgeting. It is a financial estimate of the most likely future result that the company expects based on the current assumptions about the future.
The concept aims to provide a continuous and constant planning horizon, that projects the expected outcomes (Zeller, Metzger 2013, pp. 299-305). It relays on a periodical cast of short-term budgets that usually cover a medium horizont from 6 up to 18 months. This process can for example happen quarterly or even monthly. This follows an iterative process in which each forecast will be updated within that defined frequency, with the benefit that each forecast is reviewed and replanned multiple times. This allows a company to modify the expected financial performance based on the newly available information of each iteration. An example of such a quarterly rolling forecast process can be seen below.
| Quarter | Jan-Mar | Apr-Jun | Jul-Sep | Oct-Dec | Jan-Mar | Apr-Jun | Jul-Sep | Oct-Dec | Jan-Mar |
|---|---|---|---|---|---|---|---|---|---|
| Time of forecast | |||||||||
| December | QTR 1 | QTR 2 | QTR 3 | QTR 4 | QTR 5 | QTR 6 | |||
| March | QTR 1 | QTR 2 | QTR 3 | QTR 4 | QTR 5 | QTR 6 | |||
| June | QTR 1 | QTR 2 | QTR 3 | QTR 4 | QTR 5 | QTR 6 | |||
| Semptember | QTR 1 | QTR 2 | QTR 3 | QTR 4 | QTR 5 | QTR 6 |
As the example tries to emphasize, each quarter is planned multiple times in order to steadily improve the quality of the forecasted data as well as to ensure the recency of information (Sivabalan 2011, pp. 46-47).
But this budgeting approach is not only a more forward-looking approach of budgeting, but it also establishes a new type of leadership in a company. Traditional budgeting is based on the idea of confined budgets, which sets performance expectations that are set in stone - it is a backward-looking approach. While rolling forecasting supports the idea of proactive leadership with a focus on the most likely future outcomes rather than on the observance of defined budgets. By no longer just following targets (where the company wants to go) the focus is on the forecast, which is a navigation tool for the company to know how to act and to react in order to achieve the set target. There, the time horizon of rolling forecasts fits between the operating business and the strategic planning of a business (Zeller, Metzger 2013, pp. 299-305).
Other types of forecasting
Other than the above-described rolling forecasting, there are also other forecasting methods, which are more statistically driven and also methods that are not used on corporate but on a macroeconomic level. Examples of such forecasting methods or approaches can be found below:
- Statistical and econometric models (Petropoulos et al. 2022, pp. 715-731),
- Bayesian Forecasting (Petropoulos et al. 2022, pp. 731-733),
- Data-driven methods (Petropoulos et al. 2022, pp. 739-748),
- Methods for intermittent demand (Petropoulos et al. 2022, pp. 748-750).
| Forecasting — recommended articles |
| Contribution analysis — Relevant information — Economic feasibility — Capital market theories — Continuation Pattern — Stage-Gate process — Performance indicators — Goal intensity matrix — Strategic planning functions |
References
- Fama, E. F. (1970), Efficient Capital Markets: A Review of Theory and Empirical Work, "The Journal of Finance 2/1970", pp. 383-417.
- Hyndman, R.J., & Athanasopoulos, G. (2021), Forecasting: principles and practice, 3rd edition, OTexts: Melbourne, Australia.
- Klimberg, R. K. et al. (2010), Forecasting performance measures - what are their practical meaning? Kenneth D. Lawrence und Ronald Klimberg (Ed.): "Advances in Business and Management Forecasting", Bradford: Emerald Group Publishing Limited, pp. 137-147.
- Petropoulos F. (2022), Forecasting: theory and practice, International Journal of Forecasting, Volume 38, Issue 3.
- Sivabalan, P. (2011), Start rolling, "Charter", Volume 82, Issue 8, pp. 46-47.
- Zeller, T. L., Metzger, L. M. (2013), Good Bye Traditional Budgeting, Hello Rolling Forecast: Has The Time Come? "AJBE 6 (3)", pp. 299-310.
Author: Martin Friesen
