Estimation of Audit Delay Determinants: Do Outliers and Asymptotic Properties Matter?

Abstract

The overriding objective of the study is to empirically examine if outliers and asymptotic properties of estimators matter in the estimation of audit delay determinants. The study employed the ex-post causal research design and focuses on a sample of ten (10) listed oil and gas firms in Nigeria. Secondary data from the content analysis of annual reports spanning the period 2010-2019 was used for the study. The study investigates if outliers and asymptotic properties matter in estimation outcomes comparing the following estimators; the standard OLS, Bootstrapped OLS and Robust estimators. The outcome of the study revealed that the robust estimator yields results that are significantly different from those of both the OLS and Bootstrapped OLS estimations. This suggests that the failure to address outliers in standard OLS estimations can significantly bias the estimation outcome and may be responsible for the myriad of inconclusive outcomes observed in the extant academic literature. Hence, the study confirms that in the estimation of determinants of audit delay, the considerations of outliers indeed constitute a significant statistical consideration for researchers and even more germane than asymptotic concerns.