This website uses cookies.

By clicking the "Accept" button or continuing to browse our site, you agree to first-party and session-only cookies being stored on your device to enhance site navigation and analyze site performance and traffic. For more information on our use of cookies, please see our Privacy Policy.

« Back to Results

Stationarity Tests and Margin of Error in Forensic Economics

Paper Session

Friday, Jan. 4, 2019 2:30 PM - 4:30 PM

Atlanta Marriott Marquis, International 4
Hosted By: National Association of Forensic Economics
  • Chair: William G. Brandt, Brandt Forensic Economics, LLC

Stationarity Tests on Medical Net Discount Rates

David Schap
,
College of the Holy Cross
Robert Baumann
,
College of the Holy Cross

Abstract

Medical net discount rates (MNDRs) are formulated for the post-1980 era using the various available Treasury instruments of between 3-month and 10-year duration. Net discount rate series for the medical care CPI and each of its two main subcategories (medical care commodities and medical care services) are constructed and their time series properties are examined. Stronger stationarity evidence exists for subsets of the data compared to the entire sample frame. Using a diagnostic technique to identify potentially stationary subsets, MNDRs for overall medical care costs and medical care services are found to possess strong stationary properties for series beginning in middle or late 2000 depending on the particular series. For MNDRs using medical care commodities, there are strong stationary properties for data sets beginning in late 2008. Total offset, which posits offsetting influences of the medical cost growth and discounting factors in MNDRs, is tested and yields mixed results.

JEL Code: K13 Forensic Economics.

Stationarity of Equity Risk Premiums Derived from Duff and Phelps SBBI Data

Stephen Horner
,
Economic Consultant
Steven J. Shapiro
,
New York Institute of Technology

Abstract

This paper discusses tests of time series stationarity of equity risk premiums computed with monthly data from the current edition of Duff and Phelps, Stocks, Bonds Bills and Inflation Yearbook (SBBI). The paper also discusses the methodologies used by SBBI to form portfolios of various asset classes.

JEL code: K13 Forensic Economics

The Margin of Error on Damages Calculations in Class Action Wage and Hour Cases

Jeffrey S. Petersen
,
Allman & Petersen Economics, LLC
Phillip H. Allman
,
Allman & Petersen Economics, LLC

Abstract

The margin of error is a controversial statistic when projecting class wide damages from a data sample in a class action wage and hour case. The source of controversy stems from two legal decisions that state certain margins of error on the sample mean were too high for damages to be projected from the sample data. These cases are Bell v. Farmers Insurance and Duran v. U.S. Bank. In Bell, the margin of error found unacceptable was 32.4 percent and in Duran the margin of error found unacceptable was 43.3 percent. However, neither of these cases specified the acceptable range for the margin of error. In both cases, the courts were trying to balance awarding damages when a relatively high error rate is present (Type I error) versus failing to act on a data sample that showed the class was likely due damages (Type II error). In both cases, the courts were persuaded that the possibility of a Type I error outweighed the possibility of a Type II error. This balancing act is a decision between trying to protect the interests of the defendant who might overpay for damages versus the interests of the class members who were likely entitled to some amount of compensation for unpaid wages. Moreover, these legal decisions do not address the potential remedies for a margin of error that may be considered too high such as two-tailed and one-tailed confidence intervals. These two statistical tests provide a remedy for a high margin of error by ensuring that defendants will not be overpaying for damages when they are projected from the lower bound of the confidence interval.
Discussant(s)
David Tucek
,
Value Economics, LLC
Scott Dale Gilbert
,
Southern Illinois University-Carbondale
Dwight Steward
,
EmployStats
JEL Classifications
  • K1 - Basic Areas of Law
  • C4 - Econometric and Statistical Methods: Special Topics