Chapter 4. Modeling

In the last chapter we used the past to predict the future. This works well for simple situations, but things are often more complicated. Most things depend on other things. Forecasting stock prices, credit scoring, predicting the weather, and designing a direct mail campaign all depend on independent data that influences the thing being predicted. If you want to predict tomorrow's weather in Chicago, you have to consider today's weather further west. They are connected.

In this chapter we look at using Excel to model a complex situation. We consider selecting independent data items and preparing them for use. It is not always easy to decide what value to predict, so we examine this process. Finally, we go through the steps and techniques needed to build a working model.