Forecasting - Part I: StatTools
A forecast is a planning tool that uses various methods to build predictions of the future based on past and present data. Businesses use foreacasts for determining how to allocate resources.
According to the book Practical Mangement Science by Wayne L. Winston and S. Christian Albright,, forecasting methods can be categorized into three groups: 1) judgemental methods, 2) regression methods, and 3) extrapolation methods.
In this blog, I will cover the use of extrapolation to create forecasting models for your finances. Extrapolation methods for forecasting also are referred to as “time series methods.” (Albright & Winston, 2016). Time series methods use past data to forecast future values. Therefore, as we use time series methods for finance, we can use past revenue and expenses amounts to predict future revenues and expenses. I prefer the use of time series methods because it removes any guesswork as to whether your estimates are reasonable, since the past is a great predictor of the future. Time series methods use historical patterns (such as cyclic, seasonable, and random) to extrapolate into the future (Albright & Winston, 2016).
Forecasting can be difficult, especially for small organizations just starting out. According to the article "How to Forecast Revenue and Growth" on the website Entrepreneur.com, start-ups should build forecasts with expenses first and then forecast revenues. Since your start-up will have had incurred both fixed and variable costs, you will have historical data to build your time series forecast. However, when building a time series forecast, it is important to note that time series forecasting methods do have limitations. Since such models use historical patterns to predict the future, you have to be aware that unexpected events can occur (such as natural disasters or political and economic events), which can throw off your forecast.
I use StatTools by the Palisade Corporation, a software developer that produces tools to create time series forecasts using Microsoft Excel (www.palisade.com). StatTools is an add-on to Microsoft Excel. Therefore, it is very easy to use since you use the same Excel functions. With StatTools, you can select moving average, which smooth out short-term fluctuations and highlight long-term trends, or exponential smoothing models, which also smooths out fluctuations but it give more weight to most recent activity versus older activity, to build your time series forecast. You also have the ability to test which model fits best with your data before executing the forecast.
If you need assistance building a forecasting model, please schedule a call with me at GritForce Bookkeeping here.