This research is sponsored by

This research is sponsored by the Covenant University Centre for Research, Innovation and Discovery (CUCRID).
Data In the first set of experiments, sessions lasted about two hours and subjects earned on average 18 Euros (about 24 US Dollars at the time of the experiment), including a show-up fee of 4 Euros. In the endogenous benefit treatments average earnings amounted to 25 Euros (about 27 US Dollars at the time of the experiment), including a show-up fee of 4 Euros. All sessions were conducted at the experimental laboratory of the University of Munich (MELESSA). The subject pool consisted mainly of students from a wide range of majors. Treatments were implemented using zTree [1] and subjects were recruited using ORSEE [2]. Table 1 summarizes the data provided with this Birinapant cost paper. A brief description of the variables is contained in the Excel file variable_description_pwyw_nyop.xlsx.
Experimental design, materials and methods
Acknowledgements Financial support by the Excellence Initiative of the German government through MELESSA and by the Deutsche Forschungsgemeinschaft through grants SCHM1196/5-1 & SP 702/2-1 and through CRC TRR190 is gratefully acknowledged.
Data The dataset described in this article is a long form panel dataset for monthly EMP and ρ (conversion factor) values for 139 countries, along with their associated 68% confidence interval values. EMP values are expressed in terms of percentage change in exchange rate while rho values can be interpreted as change in exchange rate associated with $1 billion of intervention by the central bank. Table 1 provides a glimpse of the EMP dataset. The data shown in Table 1 pertains to United Arab Emirates (UAE) - identified by its two letter code (AE). The two letter code can be mapped to the country name using the file “country_code_map.csv” attached with this article. The column “curr.emp” lists the monthly EMP values and column “rho” provides the value for ρ (conversion factor) for the country. For example, the value of 0.11 for January 2001 means that the UAE Dirham was under pressure to appreciate by 0.11%. The negative sign here indicates pressure to appreciate. The value of ρ of 4.82 for UAE for 2001 means a central bank purchase of a billion dollars would prevent a 4.82% appreciation in the exchange rate of UAE. The dataset also contains one standard deviation confidence intervals for the mean estimates of EMP and the conversion factor ρ. The data attached with Zygote article contains two files – “EMP_all countries.csv” contains the time series of EMP and ρ values and their associated 68% confidence intervals, and “country_code_map.csv” maps the two letter country code with their country names. The dataset has been constructed by using the methodology described in Patnaik et al. [1].
Experimental design, materials and methods To visualize the dataset, we plot the time-series of EMP values for China and India. Figs. 1 and 2 plot the monthly time-series of EMP values from 2004 to 2013 for China and India, respectively. The 68% confidence intervals for EMP values in our dataset provide a sense of accuracy of our point estimates of EMP. The confidence intervals have been estimated by using the standard errors of ρ values to simulate values of EMP. In Figs. 3 and 4 we plot the point estimates of EMP values along with their 68% confidence intervals for China and India, respectively. The construction of the EMP dataset (refer Patnaik et al. [1] for the methodology) employs data on exchange rates, GDP, trade, and foreign exchange reserves. These data have been sourced directly from Datastream. The EMP dataset has been coded in the open source language R, and parts of the code – along with the full dataset – can be accessed at: http://macrofinance.nipfp.org.in/releases/exchange_market_pressure.html
Acknowledgements The authors would like to thank Michael Hutchison, Philip Lane, Tarun Ramdorai and Rex Ghosh; and seminar participants at the 2013 LACEA Annual meetings, the NIPFP-DEA Research meetings, the International Monetary Fund, Trinity College Dublin and UC Santa Cruz for helpful comments and suggestions. We would like to thank Shekhar Hari Kumar and Vimal Balasubramaniam for excellent research assistance. The views expressed in this paper are those of the authors and do not necessarily reflect those of their affiliated institutions.