Skip to Content
Our misssion: to make the life easier for the researcher of free ebooks.

Ebook The use of Financial Ratios for Research: Problems Associated with and Recommendations for Using Large Databases

The use of financial ratio analysis for understanding and predicting the performance of privately owned business firms is gaining in importance in published research. Perhaps the major problem faced by researchers is the difficulty of obtaining an adequate sample of representative financial statements with many studies using 50 or fewer firms for analysis. However, when larger databases are used, it is important to know that they have problems as well and that adjustments to these samples must be made to permit the use of multivariate analysis techniques.

Understanding how to properly use large databases for ratio analysis will become of importance now that the Kauffman Center for Entrepreneurial Leadership (KCEL) has developed a financial statement database of more than 400,000 privately owned firms with a significant number of these including a base year and three operating years of financial statements. This database is currently available to a team of scholars working closely with the KCEL on selected internal studies. It is expected that this database will become generally available to researchers and this source of financial statement information is likely to become the standard for financial performance research in the future. For the first time, scholars will have a large commonly available database of privately owned firm financial statements that will provide direct comparisons between research findings and research formulation.

The advantage of having a large database that is readily available to researchers is easily understood. However, it is equally important to know the shortcomings and needed adjustments of large databases if meaningful research findings are to be achieved. The problems common to large databases are important and are part of almost any sample of financial statements used for financial or ratio analysis.

The need for reliable financial statement data and the importance of financial ratios for analysis and prediction is well established in the literature. Beginning with Beaver's (1966) contention that standard financial ratios can predict the financial performance of firms, many subsequent studies have attempted to demonstrate the predictive value of various techniques for estimating actual business performance.

A review of the literature describing methods and theories for evaluating and predicting financial performance reveals that although methods have become increasingly complex, few researchers adequately address the problems associated with the sample used. For example, most ratio analysis studies use multivariate analysis that is based on the assumption of a normal distribution of the financial ratios. Without confirming the approximation of normality of ratio distribution, the researcher is at risk of drawing erroneous inferences. When considering the distribution of financial ratios in any database, the normality of the distribution can be skewed by data recording errors, negative denominators and denominators approaching zero (Foster, 1986).

It appears that the more that is understood by researchers about financial performance, the greater the number of subjective and objective variables a particular predictive model is likely to include and the less attention is paid to sample weaknesses. It is common today to read about financial performance predictive techniques that are based on such things as "Inductive Learning" (Shaw and Gentry, 1988), " "Knowledge Based Decision Systems"(Stohr, 1986), "Polytomous Probit Analysis" (Dietrich and Kaplan, 1982), "Recursive Learning" (Marais, Patell and Wofson, 1984) and more recently the "Fair Issac Small Business Credit Scoring Model" (Asch, 1995).

The list of such financial performance predictive systems is extensive and each of these techniques attempts to explain and systematize the financial performance evaluation process. However, these techniques are generally referred to as "Expert Systems" (Duchessi, Shawky and Seagle, 1988) and they all combine subjective information unique to a given business as well as financial ratio analysis. Unfortunately, none of these new systems are based on error free data samples and, as a consequence, none have been proven to be reliable in terms of identifying future financial performance. In fact, evidence has been reported that indicates that the samples are not normally distributed and include inaccurate data. As such, these methods may not be of use to managers or educators as techniques for better managing businesses (Pricer and Johnson, 1996).

PDF Ebook The use of Financial Ratios for Research: Problems Associated with and Recommendations for Using Large Databases