This article is the second of four in the Unbiased Forecasting series. In this article we explore five unbiased forecast methods from the use of Excel to sophisticated analytics tools. Excel has limitations for the savvy strategic business partner, but if your toolbox only consist of Excel there are some methods that can be usefully employed, but don’t expect to become world class forecaster with only Excel in your Toolbox.
The information in the article and more will be covered at JPK Group’s 2020 Analytics Academy. Click here for more information.
Crowd forecasting
Crowd forecasting covers a method where the forecast doesn’t just rely on a single individual but a whole group of people providing their own individual views of the future. Forecasts can achieve better accuracy with a group of people rather than a single individual, as all individuals are biased. A diverse group, still with their own bias, creates a better pool of knowledge to forecast more unbiased than an individual.
The forecasts of an individual may be horribly incorrect or might be spot on, but the variability of individual performance makes it hard to know which individual to trust. But if you aggregate the forecasts of a crowd of people, you are much more likely to come up with a more accurate forecast on average.
Pros and Cons: Crowd forecasting is not unbiased, but we have included it as a relatively easy starting point in your journey to create forecast validation models. But note, the outcome is only as good as the average of the crowd’s opinion.
Competitive Intelligence
Understanding the competitive landscape has lot of value for Finance in general but can also be used as an unbiased forecast method. In all public companies quarterly announcements is a guidance forecast for the following quarters. That guidance forecast can be used to evaluate own performance. If Finance has noticed a pattern between its own performance and its competitor’s performance that insight can be used to provide an unbiased forecast
Pros and Cons: Using competitive intelligence is a way for Finance to utilize external data sources to validate own performance. The major challenge with this method is a perfect competitor doesn’t always exist.
Regression
A regression uses the historical relationship between an independent (often time) and a dependent variable such as sales, revenue, etc. to predict the future values of the dependent variable.
Pros and Cons: Regression can be performed with most tools. It also works well in Excel using a simple scatterplot and adding a regression line. This is an option if the data is relatively linear, exponential, logarithmical, etc., but the framework cannot be used if the dependent variable is seasonal. Also, note regression should only be used to forecast if the R-square is higher than 0.85.
Smoothing & Moving Average
Smoothing and Moving average covers a number of different methods including ARIMA, Holt Winter, etc. These models are statistical techniques using historical time-series data and applying algorithms to predict the future outcome.
Pros and Cons: Running these models in Excel can be very time consuming, as all models need to be run then a biased human selects the one he deems best. Another limitation using Excel is the user sets the historical time period. Different length of time will generate different forecasts. Finding the optimal historical time period to use is also a biased exercise using Excel.
Analytics tools utilizing AI have two major advantages to Excel. First, Desktop Statistical Analytics tools like Oracle Crystal Ball use AI to find the method among all the smoothing and moving average models that gives the best prediction. Second, the most sophisticated analytics tools like LightZ™ from Aurora Predictions also uses AI to pick the best historical time series that optimizes the prediction.
Desktop Statistical Tools works extremely well on high level forecasting like overall business revenue, regional revenue, etc., but has its limitation when the user wants to forecast at every dimension in the product hierarchy. For this kind of forecasting the more sophisticated analytics tools are needed.
Leading indicators
Leading indicators are industrial and economic statistics from which an indication of the value or direction of another variable (for example, a sales forecast) can be obtained. They are called “leading” because their direction or magnitude historically “leads” the focal variable. For example, we may find that unemployment rate indicates (leads) the future of a company’s revenue.
Pros and Cons: Leading Indicators can be extremely difficult to find when using Excel as your need to search though ton of industrial and economic statistics to find correlations to your performance. If a relationship is found it could change over time, so the leading indicator will no longer be leading. As such, the leading indicator forecasting works best having an analytics tool with AI to run against thousands of industrial and economic statistics to find the indicator that is leading performance.
Conclusion
The more advanced tools in your toolkit the more advanced Unbiased Forecasting you can produce. Unbiased forecasting should be used in concert with the business forecast to triangulate and validate an often stand alone business forecast. The strength of adding a variety of different unbiased methods is they are built from different data sources including crowds of people, competitive intelligence to statistics and mathematical algorithms. Where multiple forecast methods using different data sources and techniques aligns there is higher confidence to the overall forecast accuracy.
The ability to provide unbiased recommendations helps Finance play a stronger role as a strategic partner to the business. A strategic partner who can guide the business on where it overall performance will be heading in the future. The methods described in this article enables Finance to tell ‘what might happen’ but not ‘how to make it happen’. To get there Finance will need to expand its toolkit and invest in an analytics Tools that utilize Systematic Intelligence.
The next article will describe how to use Systematic Intelligence to provide an unbiased forecasts at detailed level that enable Finance to deliver deep support to their business partners including
1. How to maintain the current business – Early predictions for which customers that have high propensity of canceling their commitment
2. How to build new business – Predict which pipeline prospects that has the highest propensity of buying.
If you are interested in this topic, then join us at the next Analytics Academy!
Click here for more information and to register!
Authors
This article is a collaboration between Robert J Zwerling and Jesper H Sorensen from the organization Finance Analytics Institute (www.fainstitute.com) and is an excerpt from their book, Implementing an Analytics Culture for Data Driven Decisions – A Manifesto for Next Generation Finance. Robert and Jesper are the content creators behind the Analytics Academy and will teaching at the Academy in September!
Copyright 2019 Finance Analytics Institute, Robert J Zwerling & Jesper H Sorensen. All rights reserved. No part of this document may be reproduced without this copyright notice.