Go to the website of a forex forum, where you can find the daily data of various foreign exchange rates (http://www.global‐view.com/forex‐trading‐tools/forex‐history/index.html).
We need the close quotes of EUR/USD, USD/JPY, and EUR/JPY (click the boxes in the upper part). Then, obtain monthly data from January 2018 to June 2018 (you can download an excel file, then remove data for May, which are spurious, so you would have a 3 x 5 table).
Based on your data, (for every month) explain how much USDs we need to pay for 1 euro and 1 Japanese yen and how much EURs we need to pay for 1 JPY.
(Your answer would be (e.g.) paying 1.24 USDs for 1 EUR in January, etc.)
For every month, compute the exchange rates of EUR/JPY, USD/JPY, and EUR/USD by using the triangular parity (Do not directly copy the numbers in the given data / Use the other rates). Then, compute difference between actual (data‐given) exchange rates and parity‐implied exchange rates (i.e., actual number – implied number). In addition, compute the percentage difference of them. (i.e., (actual – implied) / actual x 100, here take the absolute values for the numerators).
Using the information based on your computation, evaluate the triangular parity. If you find that the implied exchange rates are not perfectly the same as the given data, provide the possible reasons.
From your exchange rates that are computed using the triangular parity, select a month with the largest difference. Using the U.S dollars and euro, propose an arbitrage strategy with the possible profits. How the (net) demands for EUR, JPY, and USD changes due to your arbitrage transactions? Based on the changes in demands, how the exchange rates will change?