Scientists in biotech and pharmaceutical research face a variety of challenges when analyzing data. Time and cost constraints often limit the amount of data that can be acquired. In many cases, the quality of the data makes it difficult to extract trends or estimate uncertainty. Sometimes there is too much data and identifying the most significant explanatory variables can be troublesome. Specifying which model best describes the data often requires a choice between competing models that initially appear to have similar goodness-of-fit measures.

Computational statistics provides a variety of applied quantitative methods that help solve these challenges including:

• Bootstrap – compensates for small sample sizes
• Partial Least Squares – transforms poor quality data into a useable form
• Feature Selection - identifies which variables have the most impact on a model
• Cross Validation - improves model evaluation and selection

This webinar highlights how the interactive analysis tools in MATLAB^®, Statistics and Machine Learning Toolbox™, and Curve Fitting Toolbox™ support computational statistics.

Previous knowledge of MATLAB is not required for this webinar.