Actuarial Model Outcome Optimal Fit (AMOOF) 3.0 --Free Research Software for Data Analyses and Advanced Probability Modeling

Presenter Information

James Smigaj

Document Type

Oral Presentation

Campus where you would like to present

SURC Room 202

Start Date

15-5-2014

End Date

15-5-2014

Keywords

Software, Statistics, Optimization

Abstract

The purpose of this software is to assist in the fitting of data sets using single and mixed probability density functions (pdfs), and to calculate distribution parameters and statistics regarding these curves. Nonlinear Optimization of the log-likelihood function is used to determine the best values for the pdf variables (MLE approach). Challenging measures such as Percentiles (Vale at Risk VaR) and Conditional Tail Expectations (Tail Value at Risk TVaR) for the resulting best-fitted pdf functions are found using numeric integration. AMOOF 3.0 primarily adds mixed pdfs, speed increases, interactive user interface improvements, and an additional real-time check on numeric accuracy. This software is an extension and enhancement of AMOOF 2.0, which was begun as an undergraduate software engineering project sponsored by the Actuarial Foundation.

Faculty Mentor(s)

Chueh, Yvonne

Additional Mentoring Department

Mathematics

This document is currently not available here.

Share

COinS
 
May 15th, 10:40 AM May 15th, 11:00 AM

Actuarial Model Outcome Optimal Fit (AMOOF) 3.0 --Free Research Software for Data Analyses and Advanced Probability Modeling

SURC Room 202

The purpose of this software is to assist in the fitting of data sets using single and mixed probability density functions (pdfs), and to calculate distribution parameters and statistics regarding these curves. Nonlinear Optimization of the log-likelihood function is used to determine the best values for the pdf variables (MLE approach). Challenging measures such as Percentiles (Vale at Risk VaR) and Conditional Tail Expectations (Tail Value at Risk TVaR) for the resulting best-fitted pdf functions are found using numeric integration. AMOOF 3.0 primarily adds mixed pdfs, speed increases, interactive user interface improvements, and an additional real-time check on numeric accuracy. This software is an extension and enhancement of AMOOF 2.0, which was begun as an undergraduate software engineering project sponsored by the Actuarial Foundation.