Modeling the relationship between financial indicators and company performance. An empirical study for US-listed companies.

Lukas Lorenz Höbarth

Publication: ThesisDoctoral thesis

354 Downloads (Pure)


Based on data of US-listed companies a relationship between potential indicators and company performance is to be found. To determine which variables are included in the final model, a selection process ("from general to specific") is started using the Akaike Information Criterion. A panel data analysis with fixed effect models yields the coefficients using ordinary least squares techniques. In addition, a specification test is conducted to decide whether a random effect or fixed effect model is used. Finally a binary logit model is introduced in order to predict whether each measure of firm performance will beat the average value for the firms in the market. In order to conduct statistical analysis financial data of 9,854 companies for the time period from 1986 to 2004 is obtained from the Compustat (S&P) database and hereof 1,672 audited companies are kept in the sample. Summarizing, the results indicate that companies with a low book-to-market ratio, an efficient working capital management, a small portion of liquidity, more equity and less debt, and high retained earnings for reinvesting purposes will have a better profitability performance measured by the return on investment. Further on companies with an unqualified auditor's opinion, less equity and more debt, few assets, and no retained earnings will have a better cash flow performance measured by cash dividends. Finally, companies with a low book-to-market ratio, efficient working capital, more equity and less debt, negative stock rating, few assets, high EBIT margin and high profitability will have a better market performance measured by the stock price. The results indicate that there exists a relationship between company performance and financial indicators as it was assumed in the hypothesis. Although some effects seem contrary and unreasonable (compared to previous studies) at least some of the indicators are explaining a company's performance quite well. (author's abstract)
Original languageEnglish
Awarding Institution
  • WU Vienna
Place of PublicationAugasse 2-6, A-1090 Wien, Austria
Publication statusPublished - 1 Nov 2006

Cite this