Don Lifke, Research and Development Engineer, Science and Engineering, Sandia National Labs
We all use Web-based calculators to enter information and have the tool spit out a number. For example, we can enter our vital health statistics and have our life expectancy predicted. The problem with these calculators is that we have to run them over and over when we want to consider different what-if scenarios. Wouldn't it be nice if we just had a Prediction Profiler with the model, where we could simply use the sliders to determine the predicted value? We could also visually see the effect of each variable by looking at the slope. This talk will provide examples showing how to create a custom experiment in JMP for a Web-based calculator, gather data, and create the model and corresponding Prediction Profiler. One such example will be a retirement planning tool. In addition, Monte Carlo simulations will be used to visualize the uncertainties of these predictions.