What comes to your mind when you hear the words forecasting, forecasts etc?
Invariably, you’ll think of weather forecasts. But forecasts are much more than that.
Forecasting is the process of making predictions of the future based on past and present data and analysis of trends. It’s a process that has existed for millennia, though often with dubious methodologies… Instead of looking in to a crystal ball to predict the future we are going to employ the power of statistics!
Why use forecasting?
Nearly every function within a company needs an estimate of what the future will look like as a foundation to create and update plans. For example:
• Marketers need forecasts to estimate the expected demand that will be generated from their activities.
• Salespeople use forecasts to understand what volumes are expected in each period and to evaluate the sales pipeline activity to make sure that it supports those expectations.
• Managers in the supply chain use forecasts to make timely purchasing requests, develop production plans, evaluate capacity needs, and develop logistical plans.
• Finance professionals use forecasts to prepare and understand financial plans. They also use them to report on earnings expectations.
What can be forecast?
The predictability of an event or a quantity depends the following conditions:
- How well we understand the factors that contribute to it;
- For example, in energy consumption the change in temperature will have an impact on the amount of energy we use to heat our homes.
- How much data is available;
- Generally, the more data you have to hand the more accurate your forecasts will be.
- Whether the forecasts can affect the thing we are trying to forecast.
- This is the principle of self-fulfilling forecasts. For example, if you publicly forecast a share price buyers will adjust their behaviour in order to achieve the forecasted price.
Often in forecasting, a key step is knowing when something can be forecast accurately, and when forecasts will be no better than tossing a coin. Good forecasts capture the genuine patterns and relationships which exist in the historical data, but do not replicate past events that will not occur again.
1. Average approach
a. The prediction of all future values are the mean of the past values