Prediction is basically the art and science of forecasting future events. Sometimes we take historical data and extrapolate it into the future, however there are problems with putting the past into the future as you may imagine. Sometimes we use Delphi groups or think tank personnel to use their knowledge to predict the future, and sometimes we simply rely on our intuition. At Estee Lauder we used a form of regression analysis called exponential smoothing which worked quite well for us. Stephen Jobs in a speech he gave, mentioned that we can connect the dots backwards but we cannot connect the dots forward.

As we begin our discussion on regression analysis, let’s first define independent and dependent variable. Also I want to introduce the concept of intervening variables. For example, people in France who drank wine were found to have fewer heart problems than people who did not. The conclusion was that wine is good for the heart – which I’m sure that you have heard. Later, researchers found that it was not the wine but the socializing that was going on while drinking the wine that had the healthy heart affect. We would call the socializing an intervening variable.

Offer you own definition of independent and dependent variables with an example. For you example, take a guess at a possible intervening variable that could be at play.