FIN  542                                                                                                         Prof Griffin

Spring 2003

 

Homework #1

 

NOTE: Please, provide proper explanations of your results, download the data file in Excel (HW1A and HW1B.xls).

 

Part A. (1/3)

 

·        I: Japanese Yen

 

1.      Calculate and graph a real FX rate index (with a base period of January 1972) based upon Consumer Price Indexes (CPI) for the $/¥ FX rate

 

2.      Based upon PPP, is the dollar expected to depreciate or appreciate in nominal terms vs. the yen as of August 2003?

 

·        II: Swedish Krona

 

1. In 1993 the Swedish Krona (SK) depreciated sharply vs. the $. Discuss the size of the devaluation as of January 1994 in term of PPP of the SK vs. the $.

 

III: Mexican Peso

 

1.      Calculate and graph a real FX rate index (with a base period of January 1972) based upon Consumer Price Indexes (CPI) for the Peso/$ FX rate

 

2.      According to PPP, by what percentage is the Mexican Peso overvalued/undervalued as of August 2003 (be careful about the sign and size)? In competitive terms evaluate the level of the Peso as of August 2003

 

3.      Evaluate the claim that the depreciation of the Peso in the first quarter of 1995 was a market overreaction in term of the PPP theory.

 

Part B. (2/3)

 

The spreadsheet hw1B.xls (to be downloaded from the course site) contains monthly data (recorded at the beginning of each month) on: Spot FX rates, 1-month Forward FX rates, 1-month Eurocurrency rates, Consumer price indices for a few different countries/currencies. Please, familiarize yourself with the data (check both Sheet1 and Sheet2) and make sure you understand their meaning. Then, by using regression analysis:

 

a) Test PPP for the following FX rates

            a1) Swiss Franc / US$    

a2) Mexican Peso / US$

            a3) Brazilian Real / US$

 

b) Test UIP for the following FX rates

            b1) ¥/ US$

            b2) Mexican Peso / US$

 

c) Test CIP for the following FX rates

            c1) £ / US$

 

d) Test UEH for the following FX rates

            d1) ¥ /US$

 

In all cases you will have to address the following issues:

• Are the variables suggested by the theory significantly related to the FX rate in the actual data?
• How much explanatory power does the theoretical model provide? (hint: look at the R-squared)
• If you had used the theoretical model to predict the spot FX rates on a monthly basis, how far off would have you been, on average (i.e.: what forecasting error would have you made on average across time for each FX rate?)
• Can you conclude that the theory holds true in the data to a satisfactory extent?
• For the tests in a) and b), do you observe significant differences across exchange rates? What can one learn from them?

 

While conducting your empirical tests you ought to remember that:

 

v     To run a proper regression analysis you don’t want to use the FX rate level but, rather, its rate change, along the lines we discussed in class

 

v     Interest rates are always provided in annualized and percentage terms

 

v     Inflation rates can be computed as rates of change in Consumer Price indices

v     Since PPP is formulated in terms of expected inflation rates you can either assume that the inflation rate expected for a given month is equal to the inflation rate recorded during the previous month or you can assume perfect forecasting ability for inflation:  i.e., the expected inflation rate at the beginning of each month is set equal to the actual inflation rate recorded over that month

 

v     There shouldn’t be any problem in handling missing values (#N/A) either in Excel on in Eviews (if you choose Eviews for running your regressions).  The results will simply refer to the periods for which data are available.