Unobservable parameters and conditional estimates of internal rate of return

Steven R. Fritsche, Michael T. Dugan

Research output: Contribution to journalArticle

Abstract

The conditional estimate of internal rate of return (CIRR) is intended to address several conceptual problems of the accounting rate of return (ARR). Nevertheless, researchers have raised concerns about the need to use an assumed project life and cash flow profile to calculate CIRR. The inability to directly measure IRR has prevented researchers from determining the impact of erroneous assumptions about firms' project lives and cash flow profiles. The current study addresses this gap in the research literature by using simulation techniques to observe the IRRs, project lives, and true cash flow profiles for a sample of firms. By simulating the results of operations for a sample of firms, observations of the IRR, life of the composite project, and cash flow parameter were obtained to support a detailed evaluation of CIRR. The results of the current study indicate that assuming incorrect cash flow profiles affects the error with which CIRR estimates IRR. Moreover, the nature of the effect does not appear to be consistent with expectations. The results also indicate that assuming incorrect values for the life of the firm's composite project will affect the estimation error in CIRR. The impact of erroneous assumed project lives appears to be more pronounced when CIRR is estimated for shorter sample periods. Growth did not significantly influence the estimation error in CIRR for the sample of simulated firms. The results of the current study support prior research, which suggests that sensitivity analyses may not fully compensate for the use of assumed values for unobservable parameters. Additional research is needed to identify techniques that will allow decision makers to more accurately determine the parameters needed to calculate CIRR. Also, more information is needed about the unique characteristics of the firm and its economic environment that can increase the error with which CIRR estimates IRR.

Original languageEnglish (US)
Pages (from-to)1-17
Number of pages17
JournalAcademy of Accounting and Financial Studies Journal
Volume13
Issue number2
StatePublished - 2009
Externally publishedYes

    Fingerprint

ASJC Scopus subject areas

  • Accounting
  • Finance
  • Economics and Econometrics

Cite this