Jake VanderPlas • 7/6/2015

The Model Complexity Myth

This technical article challenges the common statistical rule of thumb that models cannot have more parameters than data points. It explains the 'model complexity myth,' detailing how under-determined systems can be addressed from both frequentist and Bayesian perspectives, with practical examples and Python code demonstrating their utility in scientific applications like accounting for measurement bias.

0 comments

#Statistical Modeling #Bayesian #Linear Models