Date of Award

2020

Document Type

Dissertation - Open Access

Degree Type

Dissertation

Degree Name

Doctor of Business Administration (DBA)

Advisor(s)

Halil Kiymaz

Second Advisor

Nana Amoah

Third Advisor

Koray Simsek

Keywords

random sequence perception, time series, statistical process control charts, randomness, behavioral finance, partial least squares structural equation modeling

Abstract

The purpose of this dissertation was to determine if finance and accounting personnel could distinguish between random and non-random time-series strings and to determine what types of errors they would make. These individuals averaging 13 years of experience were unable to distinguish non-random patterns from random strings in an assessment composed of statistical process control (SPC) charts. Respondents scored no better than guessing which was also assessed with a series of true-false questions. Neither over-alternation (oscillation) nor under-alternation (trend) strategies were able to predict type I or type II error rates, i.e. illusion of control or illusion of chaos. Latent class analysis methods within partial least squares structural equation modeling (PLS-SEM) were successful in uncovering segments or groups of respondents with large explained variance and significant path models. Relationships between desirability of control, personal fear of invalidity, and error rates were more varied than expected. Yet, some segments tended to illusion of control while others to illusion of chaos. Similar effects were also observed when substituting a true-false guessing assessment for the SPC assessment with some loss of explained variance and weaker path coefficients. Respondents also provided their perceptions and thoughts of randomness for both SPC and true-false assessments.

Share

COinS