Long -Term Value at Risk: Modeling and testing VaR on a Defined-benefit Pension Scheme

Mavani, Bansi Purshottam (2020) Long -Term Value at Risk: Modeling and testing VaR on a Defined-benefit Pension Scheme. Masters thesis, Dublin, National College of Ireland.

Preview

PDF (Master of Science) Download (2MB) | Preview

Abstract

Value at Risk is identified amongst the significant developments in the field of risk measurement and management. Value at Risk (VaR) can be regarded as a metric that enables us to calculate and quantify the degree of risk of financial nature associated with a given investment or a portfolio of investments over a specific period.
The objective of this study is to compute long horizon Value at Risk on a defined-benefit pension scheme. Such pension schemes are regarded as defined in the context of the benefit formula which is well in advance defined as well as known. To accomplish the stated objective the author undertakes the building of a model for computing VaR with Monte Carlo simulation using a non-parametric bootstrapping method. The method employed for computation is not a very common one and therefore the author aims to explore its validity. An important aspect of any constructed model is check for its validity and accuracy because unless the accuracy of the model is proved the computation cannot be relied on. Backtesting of a model is performed as a test for its accuracy. To achieve the aim of building and backtesting the VaR model use of Microsoft excel along with virtual basic for application (VBA) and R programming is made. This study addresses more of an industrial problem and thereby incorporates the floors and options to various risk factors involved in the computation. The time horizon and the confidence level are the two important parameters in VaR computation. The time horizon considered for VaR computation is long-term and at 99% confidence level whereas the backtesting is performed with 95%, 97.5%, and 99% VaR. The results of the study conclude the constructed VaR model as valid and accurate for practical implementation based on the backtesting results.

Keywords: Value at Risk, Long-term horizon, Defined benefit Pension Scheme, Monte Carlo Simulation, non-parametric bootstrapping, Backtesting.

Item Type:	Thesis (Masters)
Subjects:	H Social Sciences > HG Finance H Social Sciences > HG Finance > Investment
Divisions:	School of Business > Master of Science in Finance
Depositing User:	Dan English
Date Deposited:	16 Feb 2021 09:58
Last Modified:	16 Feb 2021 09:58
URI:	https://norma.ncirl.ie/id/eprint/4765

Actions (login required)

View Item