Academic journal article The Journal of Business Forecasting

Answers to Your Forecasting Questions

Academic journal article The Journal of Business Forecasting

Answers to Your Forecasting Questions

Article excerpt

[ Q ] What is the diff erence between outlier and seasonality?

[ A ] An outlier is an unusual value that occurs once in a while. Sales of a certain department store went up signifi cantly in one month because of the Olympic games in town, or dropped sharply because of heavy snow. These are outliers that occur, but not at any set time. Seasonality, on the other hand, occurs every year regularly and at a set time. For example, sales in department stores go up in the month of December because of Christmas. This happens every year and at the same time. Sales of winter sport goods increase during winter, again this happens every year at the same time.

[ Q ] How do we determine the seasonality of a product that has a life span of just one year?

[ A ] Certainly, we need data of more than one year to make sure seasonality refl ected in the data is true seasonality. If in one year in January, sales went up sharply, but in other year in the same month it did not, then it is an outlier, not seasonality. If a product has seasonality it should consistently, year after year, do well or poorly in the same months. One way to tackle this problem is to look for its analog products and see if they do well and poorly in the same months. In other words, if a product under consideration did very well in January, and so did the analogs, then it is a seasonality. This is fi ne if the analogs were launched in the same month as the product under consideration, which may not be the case. Each product goes through the following phases: In the fi rst phase, sales increase, though very slowly; in the second phase sales increase at an increasing rate; in the third phase sales increase at a decreasing rate; and in the fourth phase, sales reach a peak and then start declining. In making this determination, make sure that the increase or decrease in sales is because of seasonality, not because of the phase the product is in. Here we may have to use our judgment.

[ Q ] What is the best way to disaggregate category sales into SKUs?

[ A ] There are several ways of doing it, the one which is often used is the ratios of SKUs to a category over the last nine or twelve months. In other words, what percentage of the sales of SKU-1 of its total category came in over the last twelve months, what percentage of sales of SKU-2 of its total category came in during the same twelve months, and so on? …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.