Our analysts turn your data into a powerful Bayesian Network Model using a Hierarchy of Causation.  

Our models are better than your analysts can build because:

  • We Consult: The Jube Capital Analyst will get a feeling for the problem you are trying to solve and suggest improvements in the data structure before the modeling process takes place.
  • We Clean: The Jube Capital Analyst will clean the data removing outliers where appropriate and / or upon request.
  • We Transform: The Jube Capital Analyst will try thousands of variable transformations and compounded variable combinations to obtain stark improvement on the correlations observed in the raw variables alone.
  • We Model: The Jube Capital Analyst determines an optimal Bayesian Network Structure comprised of node Hierarchy, Tree Augmentation and Causation Arc \ Link direction.
  • We Deploy: The Jube Capital Analyst will present a final model and advise on its implementation using our API.