When it Comes to Finance, ‘Normal’ Data is Actually Quite Uncommon

Researchers in business often rely on assumptions to analyze their findings. However, inaccuracies in these assumptions can result in serious complications—more frequently than one might anticipate. This was highlighted in a recent study that evaluated financial data from roughly a thousand major US companies.

A common assumption in data analysis is that the data will follow a normal distribution, often called the bell curve in statistics. If you’ve looked at a chart showing the heights of people, you’ve seen this curve: most individuals cluster around the average, with fewer at the extremes. Its symmetrical and predictable characteristics are often taken for granted in research.

But what happens when real-world data strays from this orderly distribution?

As business professors, we analyzed financial data from publicly traded US companies—looking at metrics like market value, market share, total assets, and other financial ratios. Researchers often investigate this data for insights into company operations and decision-making.

Our research shows that this data seldom conforms to the bell curve. In some cases, we found extreme outliers, with large firms being thousands of times bigger than their smaller peers. Additionally, we noticed distributions that are “right-skewed,” which means most data points are concentrated on the lower end of the spectrum. Essentially, while most values are smaller, a few exceptionally high figures inflate the average. This is consistent with the nature of financial metrics that are typically non-negative; for example, it’s hard to find a company reporting a negative employee count.

The Importance

If business researchers base their conclusions on flawed assumptions, their insights—such as those regarding what drives company value—could be misleading. These inaccuracies can lead to significant consequences, impacting business decisions, investor strategies, or even public policies.

Take stock returns as an example. If research assumes that these returns follow a normal distribution but they are actually skewed or littered with outliers, the conclusions drawn may be incorrect. Investors relying on this research could be misled.

Researchers understand that their work has real-world implications, which is why they often spend years fine-tuning a study, gathering feedback, and revising the article before it goes through peer review and publication. However, if they fail to check whether their data adheres to a normal distribution, they may overlook a significant flaw. This oversight can undermine even the most rigorously structured studies.

This prompts a crucial question for researchers: Do I truly understand the statistical methods I’m using? Am I critically evaluating my assumptions—or am I simply taking them at face value?

Unanswered Questions

Despite the vital role of data assumptions, many studies fail to report tests for normality. Thus, it remains unclear how many findings in finance and accounting research are built on shaky statistical foundations. Greater efforts are needed to understand how widespread these issues are and to encourage best practices in testing and addressing them.

While not every researcher needs to be a statistician, anyone dealing with data would do well to ask: How normal is it, really?

D. Brian Blank, Associate Professor of Finance, Mississippi State University and Gary F. Templeton, Professor of Management Information Systems, West Virginia University.

This article is republished from The Conversation under a Creative Commons license. Read the original article.