2. Estimation: sorts continuously valued outcomes
Using new data to predict whether a given data instance is above or below a threshold
Requires a model to determine the threshold level
Making predictions and determining churn rates
3. Prediction: similar to estimation but with expectation that there will be some check in the future
Uses a training set with historical data and a known predictor variable
Predicting the size of a balance likely to be transferred when a person accepts a credit card offer
III. Examples of web services
| Previous slide | Next slide | Back to first slide | View graphic version |