Advanced statistical techniques, such as structural equation modeling, or Bayesian modeling, or choice models such as logit or probit, are common methods that doctoral students learn about.

FORBES: CMO Challenge: Getting Value Out Of Data

For example, the very choice of model variables and model processes that are investigated are often based upon subjective judgment and experience of the modeling community.

FORBES: Climate Forecasting Models Aren't Pretty, And They Aren't Smart