Models, Bootstraps, and Monte Carlo...Oh My! Designing With Small Samples (2025-US-30MP-2503)

In the development of new products or processes, sample sizes are often small. Even so, we want to make educated decisions about the performance of the new product or process. This presentation demonstrates the use of bootstrapping, modeling, and Monte Carlo simulation to understand how a process might perform, even with small samples. Because these tools are readily available in JMP, they can easily be incorporated into any modeling analysis.

Bootstrapping is useful with small sample sizes to help understand expected variation in parameters being estimated from the sample. Once this variation is understood, modeling, such as regression analysis, can be used to identify critical factors in the system and developa  predictive model between those factors and the reponses of interest.  Finally, Monte Carlo simulation is used to understand how variation in the input factors will affect the response variables.