![member](/site_assets/img/cart.png)
![book image](/images/thumbnails/4a/4a4654be3ac86fd47f55d369b9748924.png)
WORKSHOP I
"Practical Applications of Multiple Regression in a Business Environment"
This workshop will focus on the implementation and integration of sales forecasting and marketing analysis. It is a systematic approach that combines both market theory and statistical techniques to produce structural analysis, policy evaluation, and sales forecasts all of which provide actionable decision support analysis. It is a three step process that identifies and assesses factors that make up the marketing mix for a particular product line, then via simulation choose among alternative policies, and finally, produces a sales forecast of consumer and customer demand. This multiple regression based approach offers three products, whereas intuitive qualitative forecasting methods and time series methods focus exclusively on forecasting. This approach offers the most to the analyst because it emphasizes structural analysis or causality among the market drivers in the model, which may either be single-equation or multi-equation in scope.
Director, Worldwide Sales Forecasting & Marketing Analysis
POLAROID CORPORATION
WORKSHOP II
"Improving Your Forecasts Using Time Series Methods"
This workshop is designed for anyone who needs to prepare forecasts based on historical demand. It covers all facets of time series forecasting including selecting the appropriate technique, building and diagnosing a model, and interpreting the results. Numerous live examples are presented to illustrate how the techniques are applied to real data. A special emphasis is placed on exponential smoothing models. Exponential smoothing has long been the method of choice for univariate forecasting due to its accuracy and ease of use. In addition to the standard exponential smoothing models we will also discuss two recent extensions-event models and multiple level models. Event models provide adjustments for special events like promotions, strikes or other irregular occurrences. Multiple-level models work with product hierarchies and create forecasts that are consistent at all levels and can exploit relationships between levels.
Vice President
BUSINESS FORECAST SYSTEMS