Swaption pricing under Libor market model using Monte-Carlo method with simulated annealing optimization

Simple item page
dc.contributor.authorOndieki, Kennedy Munene
dc.date.accessioned2022-06-13T08:43:32Z
dc.date.available2022-06-13T08:43:32Z
dc.date.issued2021
dc.descriptionResearch thesis submitted to Strathmore University in fulfillment of the requirements for the Master of Science in Mathematical Financeen_US
dc.description.abstractThe thesis seeks to use simulated annealing optimization to minimize the difference between the value of the libor model volatility and the ones quoted in the market for a congruent pricing of a Swaption contract. The simulated annealing optimization technique, being a global minimisation method, would provide accurate parameters that will simulate libor rates that are harmonious with the observed yield curve. This latter feature employed in a Monte- Carlo pricing method would price the Swaption contract fundamentally closer to its market value than other local optimization method. The SA method starts from an initial point, often random, then searches the neighbourhood of the current solution for the next point. The neighbourhood function search is in accordance with the set probabilistic distribution that will determine the distance between the two solution. Each solution has a cost value associated with it. The cost function determines the eligibility of the solution by measuring its discrepancy with the set limit. If the discrepancy is larger than the set limit, a new solution is sought. If the discrepancy is still large, the old and new cost value is compared, and the latter is accepted if its less than the former or otherwise rejected with a certain probability that is largely dependent on the control mechanism. The method terminates if the cost value attained is equal to the set tolerance level. Different from other heuristic methods that solely base their solution on iterative improvement of the solution’s cost value, simulated annealing accepts some inferior solutions so as to have a wider search in the design space. The main advantage of the method is the ability to escape local minimum entrapment through the aforementioned acceptance/rejection criteria. The results indicate that the advantageous aspects of the Simulated annealing enable it to outperform the least square non-linear optimization method commonly used in simulation.en_US
dc.identifier.urihttp://hdl.handle.net/11071/12819
dc.language.isoenen_US
dc.publisherStrathmore Universityen_US
dc.subjectSwaption pricingen_US
dc.subjectMonte-Carlo methoden_US
dc.subjectSimulated annealingen_US
dc.subjectAnnealing functionen_US
dc.subjectAcceptance/ rejection criterionen_US
dc.titleSwaption pricing under Libor market model using Monte-Carlo method with simulated annealing optimizationen_US
dc.typeThesisen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Swaption pricing under Libor market model using Monte-Carlo method with simulated annealing optimization.pdf
Size:
2.13 MB
Format:
Adobe Portable Document Format
Description:
Full - text thesis
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
MSc.MF Theses and Dissertations (2021)