Using Multiple Linear Regression Statistical Analysis To Calculate And Estimate Organizational Risk

As organizations are routinely bombarded with a plethora of constantly changing and complex quantitative information, analytical procedures are being utilized to absorb and make use of this information in a sophisticated fashion.  Just in the past decade, millions of business people have discovered that one of the most effective ways to analyze and evaluate decision alternatives involving electronic spreadsheets to analytically model the decisions they face.

Using spreadsheet models, a business person can effectively analyze decision alternatives before choosing a specific plan for implementation.

Spreadsheet modeling can primarily assist audit departments by enhancing their productivity through improved audit focus.

Using Regression as a structured modeling approach to decision making will produce, on average, good outcomes more frequently than making decisions in a haphazard manner.  Questions such as, "Do I need to perform more a detailed review for duplicate payments this month or is the risk of loss at an acceptable level?" can be easily answered using this approach.  In essence, one of the main objectives of modeling, as it relates to auditing, is to match the limited resources of the department to the business risks associated with not meeting business objectives.

Therefore, this article will focus on the use of analytical procedures and, more specifically, regression analysis ("R.A.") to assess those areas that are considered at most risk to the organization.  For example, regression models will be built to monitor the risk of unauthorized payments or the level of unauthorized computer accesses.

By definition, R.A. is a statistical modeling technique used to identify meaningful, stable relationships among sets of data.   Although some of the statistics utilized may be advanced for some, this article will prove that, using a tool as simple as Microsoft Excel, anyone can complete regression tests. Detailed instructions will be provided on how to complete regressions on Excel, as well as, a full list of tests that can be completed once the basic concepts are grasped