Alfonso Flores-Lagunes, William C. Horrace, and Kurt E. Schnier, "Identifying Technically Efficient Fishing Vessels: A Non-Empty, Minimal Subset Approach", Journal of Applied Econometrics, Vol. 22, No. 4, 2007, pp. 729-745. The data in this paper are confidential under the Magnuson Fishery Conservation and Management Act (Magnuson Act), and are maintained by The National Marine Fisheries Service (NMFS): Http://www.nmfs.noaa.gov Access to the data may be granted by contacting NMFS and completing a "STATEMENT OF NONDISCLOSURE FOR USING CONFIDENTIAL FISHERIES DATA." Upon authorization from NMFS, the data can be obtained directly from NMFS or from the authors. The data are from the log-books of the Northeast Atlantic herring fishery. They consist of 2894 logbook entries for 39 vessels during the years 2000 through 2003. The complete data set contains 3004 observations. However, we used a filter of 0.10 metric tons of herring landed in a trip and eliminated observations for which vessel characteristics were not recorded. In addition, two boats within the data set were sold during the time period and have been treated as four distinct vessels due to the change in ownership. In order to focus solely on herring catch, we eliminate all observations for which the herring composition did not exceed 90% of the total catch. There are three primary methods used to capture herring: mid-water trawl, purse seine, and paired mid-water trawl. Our data included only the purse seiners and mid-water trawlers. We constructed the hours fished by calculating the difference in the departure and arrival time for each trip, and subtracting steam time. Steam time was calculated by determining the distance between port and the centroid of the reported area fished and assuming a typical speed of 12 knots steaming to and from the fishing grounds. In the case that our steaming time calculations exceeded the hours spent on the trip, we set the number of hours fished to one. The total trip time consists not only of steaming time and fishing time but also search time. The distance measures were calculated using the earth model distance conversions developed by C.G. Carlson and D.E. Clay for site-specific management guidelines available at www.ppi-far.org/ssmg and programmed in MATLAB 7. The average travel speed was provided by contacts at Woods Hole Oceanographic Institute. In the case that no crew members were observed, we substituted the mean number of crew members utilized by vessel i within the data set for the missing value. There were two vessels in the data for which we did not observe any crew data, and we set ln(Crewit) equal to zero for these boats. Models were estimated using random starting points and the Constrained Maximum Likelihood (CML) algorithm in GAUSS. William Horrace whorrace [AT] maxwell.syr.edu William C. Horrace Associate Professor of Economics The Maxwell School of Syracuse University 426 Eggers Hall, Syracuse, NY 13244-1020 http://faculty.maxwell.syr.edu/whorrace/