5. Regression Analysis A powerful way of examining business trends.
We are often faced with the task of assessing whether a business suffered a downturn as the result of an event. For example, in one case we investigated, the plaintiff alleged that his business had been defamed on a television program, and as a result, sales had decreased. Another plaintiff (a tenant of a shopping centre) alleged that the shopping centre owner had ceased to properly maintain and promote the shopping centre and that as a result, sales of the plaintiff's business had fallen. Typically in these situations, annual financial statements are produced in support of allegations of loss. These may indeed show a noticeable decline in sales in the financial year following the alleged event. Case proven? Not necessarily.
When we looked closely at figures for the businesses mentioned above, we found that declines in sales had commenced prior to the alleged events. In one instance, we showed that the beginning of the decline was linked in time to the opening of competition. In both instances, we found no detectable impact on sales from the date of the alleged event.
Regression analysis is a technique we have applied. It can be used to analyse the relevance of an event through its impact on an aspect of a business eg sales. Regression analysis can be used to examine the relationship between two variables. For example, the relationship between daily temperature and water consumption may be the association of higher daily temperatures with increased water consumption. In respect of advertising expenditure and sales, regression analysis may prove that increased expenditure on advertising boosts sales. Time and sales is another such relationship.
Regression analysis can assist in identifying a trend in a business both before and after an alleged event. If the alleged event has had no impact on a business's sales, the trend both before and after the event will be the same (or at least very similar). On the other hand, if there has been an impact, the trends before and after the alleged event will differ.
In the two matters referred to above, there was a declining trend in the sales of the two subject businesses after the alleged event. However, the declining trend had started some months prior to the alleged event, and the trend in sales both before and after the alleged events were indistinguishable.
If you are faced with the task of assessing whether an alleged event has impacted on a business, regression analysis may prove helpful. Although regression analysis is not suitable in all situations, when appropriate, it is a powerful analytical tool.