What Is the Difference Between an Absolute & a Relative Measure of Forecasting Error?

What Is the Difference Between an Absolute & a Relative Measure of Forecasting Error?
••• traffic_analyzer/DigitalVision Vectors/GettyImages

One of the keys to any good investment strategy is the ability to trust forecasting, or projections about the likely future values of specific investments. Financial analysts study business models, market trends and economic events to produce forecasts that they publish in reports or sell to their clients, who are seeking information on profitable investments. But knowing how much to trust forecasts is as important as any other decision in the investment process.

Tips

  • While the absolute measure of a forecasting error is based upon numerical variations designed to determine a degree of error, a relative measure of error uses statistical variations to assess how far a given error is from the reality of the situation.

Forecasting Error Defined

Forecasting error is a term from the field of statistics that refers to the difference between an actual outcome and an earlier prediction, or forecast. A forecasting error can occur any time an individual or organization states a prediction in numerical, time-based terms. Within the context of investing, forecasting error measures the difference between how much an analysis suggested a given investment product would be worth on a given date and the actual value of that same product on that actual date. Ultimately, whatever form it takes, forecasting error is a measure of the accuracy of a past prediction.

Understanding Absolute Measures

An absolute measure of forecasting error is a measurement that uses numerical variations in investment forecasting to determine the degree of error. For example, if an analyst forecasts that a company's stock will be worth $90 per share by the end of the fiscal year, and the actual stock price at the end of the year is $85, the absolute measure of forecasting error if $5 per share. If another analyst forecast a value of $80, this forecast has the same $5 absolute measure of error; absolute measures take the form of positive numbers, regardless of whether they represent high or low estimations.

Evaluating Relative Measures

Relative measures of forecasting error are the major alternative to absolute measures. They use statistical variations based on percentages to determine how far from reality a forecast is. For example, a $5 absolute measure of error represents only a 5 percent relative measure of error for an investment with an actual value of $100. However, the same $5 absolute measure of error represents a 25 percent relative error if the investment product is only worth $25.

Making Investment Decisions

Both absolute and relative measures of forecasting error can help investors analyze the investment advice they receive. An adviser or analyst with a low relative measure of forecasting error makes generally accurate forecasts compared to an analyst with a higher relative measure. However, for individual investments, absolute measures of error show exactly how close, in dollar terms, a past forecast came to predicting changes in value over time.