| Custom Modeling |
|
Skillfully melding the “art” and “science” of model development and deployment.
ALI has been developing powerful custom models and delivering on analytic solutions that minimize risk, maximize profit and improve process effectiveness for over a decade. ALI custom models help businesses take proactive, profit-focused actions across the account lifecycle in such areas as account acquisition, account management, credit line management, risk, fraud and collections. In today’s competitive environment, businesses cannot afford to let customer data be under utilized. The ALI analytics team excels in finding new ways to leverage traditional and non-traditional data sources to provide strong analytic solutions, including those areas in which analytic solutions have not typically been applied. ALI's analytic services team will engage with the client's team to define the statistical model objectives, performance definitions and key success metrics and clearly identify project requirements, tasks and objectives for the development and delivery of the models. Maximizing the value of a given set of data sources is imperative to building the most effective behavioral predictors. Using state-of-the-art techniques embodied in data processing and variable generation ensures a consistent and thorough exploration of the data sources. In addition, ALI has built a core library of analytic techniques that are embedded in ALI modeling tools and are a standard part of all modeling efforts. A dedicated Research and Development staff provides innovative routines and techniques for extracting information value from data. ALI blends this “science” of model development with the “art” of first hand domain expertise to create substantially stronger models than those using traditional approaches alone. ALI utilizes a variety of analytic techniques coupled with proprietary automated modeling and rules optimization capabilities to produce a more sophisticated solution that can effectively predict dual, multiple or continuous outcomes. Logistic regression is often used to allow the clients to shape the models to meet palatability constraints. This is especially useful in ensuring the models remain intuitive to the end user and are particularly suited for creating reason codes to assist in the policy decisioning and execution such as declination of credit, administer potential adverse actions, determining optimal treatment, proactive risk mitigation tactics, etc. Other tools and processes, like Optimal Rules Discovery, use multiple analytic techniques to identify the rule set that yields the best results for the business and maximizes the net benefit the user receives from deploying an analytic solution. For more information about custom modeling please email This e-mail address is being protected from spambots. You need JavaScript enabled to view it . ![]() |
