Data:

16.1. Bonds

In this paper we use three different types of "main" data. The Danish Treasury bond

"Statsobligationer", the Danish interbank borrowing rate "CIBOR" and different stock prices and indexes.

The Danish Treasury bonds, named Statsobligationer, are retrieved from the Bloomberg Terminal at the Copenhagen Business School library. They come in different maturities, of which the most dominant is the 10-year maturity when measured on volumes traded. The information we were able to gather from Bloomberg is significantly fragmented, and several important days had missing values for a number of the different maturities. We ended up

56 deciding to only use 1-, 2-, 5-, 10- and 30-year maturities, as they were the maturities with the most observations available. The total number of possible event observations using Treasury bond data is 89. We, in some of the models, add additional data to the 2-, 5- and 10 year Treasury bond rates up to a maximum of 109 observations. We were able to get additional data on these three maturities from the Statistical Bank of the Danmarks Nationalbank.¹⁷ The interbank lending rate, CIBOR, is released every bank day at 11am. It is an average of the interest rates that the largest banks in Denmark are willing to lend to another bank, without collateral, for a specified maturity. In our dataset, the maturities range between three to twelve months. The data is retrieved from Nasdaq who get the data from Finansrådet every day.¹⁸¹⁹

The CIBOR data is quite intact and we could use all the different maturities on all the days we were interested in examining. The total number of observations given event days and data availability amounted to 122 observations.

16.2. Equity prices

The stock prices are, beside Bloomberg Terminal, also retrieved from Nasdaq, who operates the Copenhagen Stock Exchange.

OMX Copenhagen 20 index, in short C20, is used throughout this paper. It is a representation of the value of the 20 largest Danish public companies. A new index has recently been created called C20 Cap. Due to being established in the middle of 2011, it is not a very useful as a representation of the value of Danish stocks in our analysis, since our data on interest rates go back to 2005. We therefore chose to use the "older" C20 to represent the value of the largest Danish companies. This has the disadvantage that it is heavily based on the share price of the largest company in Denmark, Novo Nordisk A/S. Due to its incredible growth over the last decade, it has come to be so valuable that it affect the evolution of the C20 unrepresentatively much. We try to minimize this problem by comparing to other stock indexes such as Midcap and SmallCap, and by accounting for the relationship between interest rates and the pharmaceutical business. However, it is still an obvious disadvantage.

17http://nationalbanken.statistikbank.dk/statbank5a/default.asp?w=1920

18http://www.finansraadet.dk/Tal--Fakta/Pages/satser/regler-for-fastlaeggelse-af-cibor/dagens-satser.aspx

19http://www.nasdaqomxnordic.com/

57 16.3. Events

The events used are listed in the Appendix 3. They all constitute available information gathered from the ECB, Federal Fund Reserve and Danmarks Nationalbank. The events based on information gathered from the US are not included in the full model. However, all other events are. The total number of events is equivalent to the full number of CIBOR observations 122, since there was no reduction due to data unavailability in the CIBOR model. They have been gathered from different sources. The ECB event days are what we consider to be the most important announcements to be found in press conferences, public statements etc., all found on ECB's own webpage.²⁰Danmarks Nationalbank events include all interest changes, speeches and all quarterly press releases. This data has been gathered from the Danmarks Nationalbanks' own website.²¹²² The Federal Reserve events are all press releases regarding its unconventional monetary policies.²³

16.4. Miscellaneous

In addition to these three types of main data used in our result part, we have used additional data from various sources for creating models throughout the assignment. These are:

• Equity prices from Yahoo Finance.²⁴

• Numbers on export and GDP from Denmark's Statistical bank.²⁵

• Equity prices and specifications from Google Finance.²⁶

• Data on the Danish trade balance and trade partners from IMF.²⁷ 16.5. Limitations:

There were some gaps in the completeness of the data obtained using the Bloomberg Terminal. We realize that this might have an effect on our results. Missing events with a large effect on the Treasury bond rates could skew our results and thereby interpretation. We managed to get data throughout the entire period of interest, which leads us to believe that despite missing potentially important events, the results remain useful. Another factor that would support this claim is the fact that Wright (2012) had 28 observations; Bernanke &

20http://www.ecb.europa.eu/press/html/index.en.html

21http://www.nationalbanken.dk/en/pressroom/Pages/default.aspx

22http://www.nationalbanken.dk/en/statistics/Pages/default.aspx

23http://www.federalreserve.gov/monetarypolicy/default.htm

24http://finance.yahoo.com/

25http://www.dst.dk/da/Statistik#6

26http://www.google.com/finance

27https://www.imf.org/external/pubs/nft/2014/areaers/ar2014.pdf

58 Kuttner (2005) used 131 observations, while we have 122. Wright (2012) based his events on a test for days with large heteroscedasticity. He was able to do this because he had access to intraday data. We believe that the fact that we incorporate 122 observations should be enough to mitigate any effect that lacking events would pose. Furthermore, intraday data would have made us able to better isolate the causal effects between equity prices and monetary policy, mainly because the larger the lag, the harder it is to isolate cause and response. This is especially true when considering the fact that Denmark is a small and open economy that is influenced by many factors worldwide. Moreover, we have not accounted in our model for the general economic environment or inflation adjusted data despite adding recession dummies. Inflation is not EU-wide, and varies from country to country, which could create a difference when comparing bonds from different countries. However, since inflation has remained relatively low through the period of research, we believe the effect to be negligible. We also did not differentiate between expected and unexpected information as Bernanke & Kuttner (2005) did. We do, however, include a chapter that discusses outliers and their causal relationship to the equity prices. The outliers that created a shock to the Treasury bonds rates of ±1,5 standard deviations should, in theory, equate to unanticipated events.