"As far as the laws of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality." Albert Einstein


  • Experimental Voting Data experimental data for Levine, D. K. and T. R. Palfrey [2005]: "The Paradox of Voter Participation: A Laboratory Study," also STATA file with data definitions.
  • Book Revenue Data 2003 data on book revenues in MS Excel format. The list of books was compiled by using the query "subject: fiction and pubdate: during 9-2003 and binding: Hardcover" on amazon's power search to get a list of all fiction on amazon published during 9-2003. Of the 1913 titles listed, 1713 were in stock, had prices listed and were published in the correct month. The ISBN numbers for these 1713 books are recorded on the spreadsheet and the prices are those listed on Amazon. The total sales are through 11-2004 and represent sales reported by Ingram through their voice reporting system at 615-213-6803. The data for September 2003 was collected during November 21 - December 4, 2004, while the September 2004 data was collected during November 14-27, 2004. 2004. Sales data was available for 1284 of the titles. According to the Yahoo profile Ingram's 2003 revenue was $2.2 billion. According to Association of American Publishers total book revenue for 2003 was $23.4 billion, which would mean Ingram represents roughly 10% of the market. According to Google "answers" Ingram's market share is roughly 1/6th, a somewhat higher market share. The 2004 data was generated the same way as the 2003 data, but using publications in 9-2004. For 9-2004 there were 1388 titles listed 1235 were in stock and in the correct month, and 1153 had sales data. Note that the results of the Amazon query is not identical every time the query is run, and that the sales data except for the year 2003 is also subject to change over time, so it is impossible to exactly replicate this data. However the variation due to these changes is small. In searching for ISBNs, bear in mind that they are 10 digit numbers without commas; the spreadsheet inserts commas and suppresses the leading zeroes. Data collected by Hengjie Ai, University of Minnesota.
  • Copyright Time Series U.S. Population and copyright registrations. Sources are documented in the file.
  • Patent Time Series NSF data, sources documented inside the file.
  • R&D Cross Section IPR data was provided by Walter Park. Data as a csv suitable for use with statistical packages, spreadsheet with data sources documented, as well as list of which countries and years correspond to which observations. GDP data is here.
  • Programs for Self Control, Risk Aversion and the Allais Paradox here.