What is Predictive Analysis

Our need to understand past events has led to a discipline that we now call business intelligence. It allows us to make decisions based on statistics obtained from historical data. Descriptive analytics goes a long way in allowing sound business decisions based on facts, not feelings. However, descriptive analytics is simply not enough. In the society we live in today, it is imperative that decisions will be highly accurate and repeatable. For this, companies are using predictive analytics to literally tap into the future and define future business decisions and processes.

Application of Predictive Analysis

  • Health Care

    The use of predictive analytics is to implement effective preventive care. By knowing which patients are at a higher risk of developing a certain disease, we may put preventive measures in place to mitigate risk and ultimately save lives. Lately, predictive analytics have been the center of attention on a highly publicized contest in which historical claims data is used to reduce the number of hospital readmissions
  • Outstanding Claims

    Using predictive analysis to develop a model to predict the estimated claim amount based on the damaged part in the vehicle and replaced parts estimated cost. Company historical claims data should improve employee experience in estimating the cost of the claim , and monitor the development of outstanding claims periodically. Predictive analysis helps the company forecast its current and future liabilities with more accurate results.
  • Underwriting Predictive Modeling

    Predictive analytics is used to leverage models that predict underwriting decisions based on historical and current data. insurance industry was looking for ways to incorporate/improve the use of this data science to find ways to make the underwriting process faster, more economical, more efficient, and more consistent. Recently, new types of data are being produced and collected in insurance industry that helps better estimation the risk and related premium. Some of these new types of data include:
    • Electronic health records,
    • Smartphone data
    • Patient history reocrds
    • Public available data (medical research, government published statistics)

    While collecting this data gives the industry new opportunities to collect information, it also brings new challenges on how this data can be verified and used. Insurers need to cut down the time required to issue policies and the turnaround time due to fewer underwriting requirements. Out solution will provide the underwriter with a percentage score per policy upon the company underwriting history patterns and the company underwriting guidelines
  • Credit Approval

    Credit Analysis demonstrates several analytic techniques to examine bank decision to approve or deny credit applications. Predictive analysis will combine a logarithmic model to calculate the credit risk. This model will be able to the predict credit limit each applications should have, the factors affecting the credit approval decision and a the percentage score of accuracy for each proposed credit application.
    Using predictive analysis, banks will have two forms of predictive modeling in credit risk application.
    • Examining conventional underwriting evidence in new ways to improve risk classification and fault projection.
    • Using unconventional underwriting evidence to classify applications quickly and/or cheaply while replicating the decisions reached with conventional evidence.