Statistical Tools Help To Manage Accounts


Attendees of an NACM audio teleconference, April 25, 2007 got an opportunity to learn about key statistical concepts and tools that can provide a way to analyze and manage the financial accounts of companies. The teleconference, “Statistical Tools for Managing Accounts,” was presented by Jack Williams, a professor at Georgia State University College of Law, a director in the Financial Recovery Services Group at BDO Seidman, LLP and the Scholar in Residence at the Association of Insolvency and Restructuring Advisors.

“All data tells a story,” Williams said. He pointed out that statistical analysis is a way to determine trends and glean other information from a set of financial data. “Statistical techniques can help get a better prospective of customers,” he said. With statistics, he noted, customers could be evaluated with other like customers or with the industry as a whole. Williams pointed out that conducting statistical analyses does not necessarily require having a statistical background, just an understanding of the mathematical concepts they present. Statistics can measure central tendencies and the variability of different financial values. “Statistical tools are the steady hand maidens to the credit profession,” Williams said. “They are efficient and they are reliable.” However, he pointed out, “They won’t displace your discretion or your gut feeling.” “All professions use statistics to manage uncertainty,” he added.

Financial statement analysis can present the true picture of a company’s liquidity, it’s debt and it’s financial performance. Statistical ratios are a set of tools designed to help present this picture. For liquidity there are the current ratio, quick ratio, payables deferral period, working capital and short pays. For debt ratios there are debt to equity and times interest earned. The performance ratios include DSO period, inventory carry period, profit margin, ROA and ROE. Williams explained each of these ratios. “My research shows that short pays are a good indicator of liquidity problems,” he said. “All of those ratios are designed to show us if we’re going to get paid.”

Williams explained various statistical measures of averages that are used to summarize data—the mean, median and mode. “When we summarize data we lose some specificity, but it becomes more manageable.” The data and what one wants to get from their analysis determine what kind of average to use. The mean can be presented on an arithmetic, trimmed or weighted basis. He noted that a trimmed mean is one that discards outlying values in a data set. A weighted average is one that assigns a higher value to certain data elements. A median is the value where an equal number of data points are of higher and lower value. “Generally you should be going to the mean unless you have good evidence to use something else.” Williams also explained measures of variability in data such as range, quartile and percentile and standard deviation. He defined standard deviation as the dispersion of values around the mean. He then used some data sets to present and explain various types of statistical calculations and how they may be analyzed. These are the same methods that would be employed with data found on financial statements. He explained how to interpret the results.

In response to a question from a teleconference attendee on how to statistically evaluate a seasonal business, Williams advised splitting the financial data into “on” and “off” cycles and compare past years’ on cycles to each other and past years’ off cycles to each other. “Statistics can only help in apples to apples comparisons,” he said.

NACM teleconferences are an economical and convenient way to learn from the experts about the important topics many credit professionals must address. To find out about upcoming NACM audio teleconferences, go to NACM’s website at On the top of the page, place your cursor on the “Education” tab, then scroll down to “teleconferences.”

Source: Tom Diana, NACM staff writer

Leave a Reply

Your email address will not be published. Required fields are marked *