One of the most challenging aspects of applied predictive analytics is quantifying the impact of model outputs – both prospectively and retrospectively.
An oft-asked question is,
“How do you measure the impact of something that didn’t happen?”
In this webinar, Bret will describe methods observed “in the wild” by carriers seeking to assign hard dollars to predictive model impact.
Specific scenarios will be illustrated:
- Evaluating a pricing model: Is it better than what we currently have?
- Forecasting the impact of a pricing model: what strategy should we employ to maximize profit?
- Measuring the impact of a pricing model: what benefit did we actually accomplish?
- Projecting the impact of a prescriptive model: how much improvement is available?
- Measuring the impact of a prescriptive model: how much improvement did we actually accomplish?