Exchange Rate Volatility – Project topics materials

Get Complete Project Material File(s) Now! »

Data, Variables, and Descriptive Statistics

In this section, we present the data set that is used in the multiple time series analyses of Section 5, how we transform it, and any irregularities in the data. We discuss how we construct the variables that are used in our empirical model. These are exchange rate volatility, trade volume, price level ratio, and output. We deal with the various advantages and disadvantages of the proxies used and justify our usage of them. First, we discuss the data set and the method of calculating the variables, and we end this section with a short discussion of the descriptive statistics.

Data

We use monthly data for the period from February 1995 to October 2011, which gives us 201 observations per time series. The period covered is different for China and Belgium because not all data are available for the missing time periods. Belgium is covered from January 1998 until October 2011 (165 observations), whereas China is covered from February 1995 until November 2003 (124 observations).
We use natural logarithms on some of our variables to obtain elasticities directly from our estimated regression coefficients. In order to avoid non-stationarity for all variables, we take the first difference for all our variables except for our volatility measure. This is necessary because non-stationarity can lead to spurious regressions (Gujarati & Porter, 2009). The first difference is also taken from the natural logarithms to approximate percentage changes. If extreme outliers are identified that might negatively affect our regressions, we bind their values to the 1st and 99th percentile respectively.
The monthly average bilateral exchange rates from which we calculate the exchange rate volatility are obtained from the database of the Riksbank. Statistics Sweden (SCB) supplies the data necessary for the calculation of the price level ratio and the trade volume. The OECD provides us with the data for the output variable. The next subsection shows how we transform these data sets to construct the variables. We begin with the dependent vari-able: trade volume.

READ The entrepreneur’s bricolage strategies

Variables Trade Volume

In order to measure the impact of exchange rate volatility on bilateral trade volumes, we need to quantify the Swedish exports (EX) to and imports (IM) from the various countries under investigation. In this paper we utilise, the volumes of exports and imports, measured in metric tons. Our data for exports and imports only include the volume of goods traded and disregard the bilateral trade in services. Later, this narrows down our analysis and has to be kept in mind. We do not use the absolute volume but rather the growth in the volume traded ( ).
This is accomplished by taking the first difference of the natural logarithm of the exports and imports respectively:

Introduction
2 Background
2.1 Exchange Rate Volatility
2.2 Previous Studies
3 Theoretical Frameworks ..
3.1 Hooper and Kohlhagen .
3.2 De Grauwe .
3.3 Broll and Eckwert
4 Data, Variables, and Descriptive Statistic
4.1 Data
4.2 Variables
4.3 Descriptive Statistics .
5 Empirical Model and Analysis .
6 Conclusions
6.1 Suggestions for Further Research
List of References