Product details

Kirk Nelson is trying to determine if a new style of girl's clothing for the Wal-Mart account is cannibalizing an existing profitable style of the same product. He has sales data for over three years on the existing style, and a year's data on the new style. However, the data has both high seasonality and upward trend, making it difficult to analyze. In addition, another new style has appeared in the last few months that may also be affecting sales. Finally, if Kirk can model the effects of these style changes in order to make a decision on which styles to keep and which to replace, he wonders if he can use the model for forecasting future sales. The case is set up for use at either a basic level, such as a first statistics course, or at an advanced level, such as in a quantitative marketing course, an operations management course, an advanced statistics course, or an MBA course in any of these areas. The teaching note includes an extensive analysis of the data, including removing the seasonality, determining the trend, making basic statistical inferences, and exploring the insights from both simple regression as well as multiple regression analyses. Various possible models are constructed to help Kirk understand how the different new styles are affecting the sales of the existing style, and then a discussion of the value and danger of forecasting from such models follows. Finally, Kirk's options, given the model findings, are detailed and their pros and cons are evaluated, both at the basic level and again for the advanced level.