Matched Pairs Discrimination: Methodology and an Investigation of Corporate Accounting Policies

Abstract

The matched-pairs methodology is becoming increasingly popular as a means of controlling atmnmus factors in business research. This paper develops discriminant procedures for matched data and a a m i n a the properties of thae methods. Data from a recent study by Hunt 1141 on the determinants of inventory method choice are used to contrast the performance of the different methods. While all of the methods yield the same m of discriminating variables. thox procedures that allow for the dependence among observations within a pair provide greater classificatory power than traditional multimriatc techniques. Subject A m Economlc Analysis, Fhmrchl Reporting, L h a v Statistical Mod& and S~rirticul 'Rchniques.

Neural network detection of management fraud using published financial data

“…Using a matched pair design has many benefits (Hunt and Ord, 1988;Sibley and Burch, 1979). An important benefit of matched pairs is that it controls for external factors.…”

An Examination of Debt‐equity Proxies vs. Actual Debt Covenant Restrictions in Accounting Choice Studies

“…This paper is meant to provide a step in the direction envisioned by these observations by examining the relative performance of the most commonly used proxy in studies of accounting choice, the debt-equity ratio (Watts and Zimmerman, 1986; and Duke and Hunt, 1990), versus measures of the actual existence and tightness of restrictive debt covenant provisions. This examination is performed by incorporating actual debt covenant data into Hunt's (1985) study of inventory method choice that used the debt-equity ratio as a proxy for accounting-based covenant restrictions. In general, our results indicate that the use of actual covenant data (both the existence and tightness of covenant restrictions) is superior to simply using a debt-equity proxy in a model of inventory method choice.…”

“…Third, Hunt's complete set of data was available to the current authors. Hunt (1985) examined six variables hypothesized to be related to the inventory method choice over the 1974-75 period. These included a dummy for the existence of an accounting-based management bonus plan (to test the bonus plan hypothesis), the level of management ownership of the firm, and four debtcovenant related variables (leverage, interest coverage, tightness of dividend constraints, and current ratio).…”

“…For purposes of the current study, several minor modifications were made to the Hunt (1985) data set, both to improve on possible weaknesses in the earlier study, and to promote comparability after actual covenant information is introduced into the analyses. First, two pairs of sample firms included in the Hunt study were eliminated due to our inability to locate covenant information for one firm in each pair.…”

“…A few studies have included some covenant restrictions in models of accounting choice (e.g., Bowen, Noreen and Lacey, 1981;Hunt, 1985;Ayres, 1986;Wong 1988;and Healy and Palepu, 1990), examinations of the economic consequences of mandated accounting changes (e.g., Lys, 1984;Frost and Bernard, 1989;Salatka, 1989;and Gopalakrishnan and Sugrue, 1992), an investigation of overfunded defined benefit pension plans (Haw, Jung and Lilien, 1991), and a study of operating decisions (Bouillon and Shane, 1990).3…”