Predictive ability of FX-rates, during the COVID-19 pandemic, using GARCH-models.
2022 (English)Independent thesis Advanced level (degree of Master (One Year)), 10 credits / 15 HE credits
Student thesis
Abstract [en]
In the beginning of 2020, the world was struck by a new disease called the Coronavirus (COVID-19), which later on was declared a pandemic. As the pandemic caused more and more infections and casualties as an effect currency, stock markets, and various other assets depreciated. Previous research has made it clear that cases and deaths related to the disease have a significant impact on both returns and volatility of FX rates. Although, there is a lack of research that just takes pandemic-related data into account, with research focusing on monetary variables such as interest rates and inflation. The purpose of this paper is to investigate whether cases and deaths in Sweden related to the disease can be used as a forecasting tool for the return and volatility of FX rates. Furthermore, the FX rates have all had the Swedish krona (SEK) in relation to other currencies as the focal point since there is a lack of research regarding the SEK. The study investigates 545 daily observations of nine bilateral nominal FX rates plus cases and deaths in Sweden related to COVID-19, with the time period ranging from March 2, 2020, to March 31, 2022. The study adopts a GARCH (1,1) and GARCH (5,5) model where, in the first model, cases and deaths are lagged one period to see the one-day ahead forecasting accuracy. The model is then repeated but cases and deaths are lagged five periods to see the five-day ahead forecasting accuracy. The results from the empirical work find that, when deaths are lagged in five periods, the forecasting ability is superior to that of cases and deaths when only lagged one period. For deaths lagged five periods, there is a significant forecasting ability for the variance of SEK/BRL, SEK/GBP, and SEK/JPY. There is also a significant forecasting ability for the return of SEK/BRL when deaths are lagged for five periods. For cases that lagged one period, there is only a significant forecasting ability for the variance of SEK/USD.
Place, publisher, year, edition, pages
2022. , p. 28
Keywords [en]
FX-rates, COVID-19, cases, deaths, GARCH, forecasting, lags.
National Category
Economics
Identifiers
URN: urn:nbn:se:hv:diva-18634Local ID: EXF800OAI: oai:DiVA.org:hv-18634DiVA, id: diva2:1674568
Subject / course
Nationalekonomi
Educational program
Magister i finans
Supervisors
Examiners
2022-06-222022-06-222022-06-22Bibliographically approved