ARMA-GARCH time series model formodelling short interest rates
Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Interest rates play a vital role in an economy. It is the main indicator of investors 'mood and views of the future. Understanding this process and ability to model it correctly can help to minimize risks when making operations on fixed income markets.
This thesis investigates the performance of ARMA-GARCH model applied to daily observations of interest rates. The work firstly investigates the main features of interest rates and shows the suitability of ARMA-GARCH framework for modelling interest rates.
The effectiveness is assessed by analyzing properties of simulated data as well as pricing contingent claims. In contrast to equity derivatives, both the derivative payoff and discount factor depend on the level of interest rates.
The findings conclude that proposed model appeared to be superior to chosen benchmark models CIR and CKLS. The simulated data has properties identical to historical and accuracy of pricing bonds and bond options is higher. Such a framework does not exclude the negative interest rates, though, and as seen recently on the markets, it is not unusual in practice. Such an approach explores a room for future research. Multivariate analysis with growth rates and inflation in time series framework is very promising.
Place, publisher, year, edition, pages
2013. , 54 p.
Interest rates, time series, ARMA-GARCH, derivative pricing
Economics and Business
IdentifiersURN: urn:nbn:se:kth:diva-142910OAI: oai:DiVA.org:kth-142910DiVA: diva2:704890
Koski, TimoZhang, Xin