Cognalysis MultiRate™ is a powerful predictive modeling tool tool for actuaries, analysts, and others whose jobs depend on understanding predictive relationships in their data.
Variable Gradient Geospatial Smoothing unleashes the predictive power of physical proximity. This method takes advantage of neighboring data points in a way that is more responsive where possible, yet more stable where necessary.
“We found that Cognalysis MultiRate provides a valuable non-distributional path to predictive analytics that’s technically sound and understandable. It supports our work and has shortened our implementation timeline.”
Jane Jasper-Krumrie, FCAS
Software Highlights
Powerful – Quickly identifies key predictive variables, separating noise from underlying relationships. Includes techniques for identifying interaction effects and refining, improving, and simplifying models automatically.
Easy to use – You do not need to be a statistician to get powerful, robust results with this software. For those that are already statisticians/data miners, add this powerful tool and compare the results to your other machine learning/GLM results.
Flexible – This tool can be applied in a wide variety of everyday analysis problems, and is capable of handling large numbers of variables. As you use the tool more, you will think of even more ways to use it.
Cost effective – Only $5,000/year for a single user; volume discounts for multiple users.
Insurance applications:
Establishing/Adjusting Rates and Rating Models
Business Strategy
Investigating Case Reserve Adequacy Changes
Parameterizing an Individual Claim Life Cycle Model
Reserve Analysis Segmentation
Price Monitoring
Fraud Detection
Audit of Underwriters
Improving Marketing Efficiency
Understanding Differences Between Catastrophe Models
many more…
For more information on how you can use Cognalysis MultiRate™ for a wide variety of analytical problems, please contact Chris Gross at (651) 293-8008 or chris.gross@cognalysis.com.
“MultiRate’s multivariate analysis capabilities have helped us better understand the primary risk characteristics driving loss experience and create a powerful new predictive model. Going forward, we expect this tool to provide significant value in pricing and underwriting our accounts.”
Understand how to compare alternative models to each other as well as how comparisons are made between the predictive characteristics that make up a model.