Radu Tunaru - Academia.edu (original) (raw)

Papers by Radu Tunaru

Encyclopedia of Financial Models, 2012

… Business School Discussion Paper No. 6, …, 2000

This paper presents a model that measures the impact of political risk on portfolio investment wh... more This paper presents a model that measures the impact of political risk on portfolio investment when the political risks are multivariate a nd c orrelated across countries. The multivari ate approach generalizes the single country model but retains most of its characteristics in t erms of its ab ility t o price political risk based on the stochastic process of

The aim of this paper is to develop a model for analyzing multiple response models for count data... more The aim of this paper is to develop a model for analyzing multiple response models for count data and that may take into account complex cor- relation structures. The model is specified hierarchically in several layers and can be used for sparse data as it is shown in the second part of the paper. It is a discrete multivariate response

Progress in Analysis and its Applications - Proceedings of the 7th International ISAAC Congress, 2010

ABSTRACT A general method for generating approximation algorithms for integral calculations is pr... more ABSTRACT A general method for generating approximation algorithms for integral calculations is proposed here, starting from weak convergence results and then adjusting the integral calculations such that the Gaussian probability kernel appears inside the integral. While this technique can be applied in a wide applied mathematical context we focus here on European option pricing as a class of applications. We prove that the weak convergence characterizing condition can still be applied under some mild assumption on the payoff function of financial options. It is also shown that the approximation grid is a dense set in the set of real numbers.

Stochastic Analysis 2010, 2010

Quantitative financial calculus is dominated by calculations of integrals related to various mome... more Quantitative financial calculus is dominated by calculations of integrals related to various moments of probability distributions used for modelling. Here, we develop a general technique that facilitates the numerical calculations of options, prices for the difficult case of multi-assets, for the majority of European payoff contracts. The algorithms proposed here rely on known weak convergence results, hence making use of the gaussian probability kernel even when modelling with non-gaussian distributions. In addition, this technique can be employed for calculating greek parameters. We prove that the weak convergence characterizing condition can still be applied under some mild assumption on the payoff function of financial options.

Journal of the Royal Statistical Society: Series D (The Statistician), 2001

IMA Journal of Management Mathematics, 2000

ABSTRACT

Journal of Portfolio Management, 2009

... The underwriter of a balance guaranteed swap is exposed to the amortization speed on the refe... more ... The underwriter of a balance guaranteed swap is exposed to the amortization speed on the reference mortgage loan portfolio. View larger version (15K). Exhibit 1. ... Jim Clayton, S. Michael Giliberto, Jacques N Gordon, Susan Hudson-Wilson, Frank J Fabozzi, Youguo Liang. ...

Index-linked securities are offered by banks, financial institutions and building societies to in... more Index-linked securities are offered by banks, financial institutions and building societies to investors looking for downside risk protection whilst still providing upside equity index participation. This article explores how reverse cliquet options can be integrated into the structure of a guaranteed principal bond. Pricing problems are discussed under the standard Black-Scholes model and under the constant-elasticity-of-variance model. Forward start options are the main element of this structure and new closed formulae are obtained for these options under the latter model. Risk management issues are also discussed. An example is described showing how this structure can be implemented and how the financial engineer may forecast the coupon payment that will be made to investors buying this product without exposing the issuing institution to risk of loss.

The departure from the log-normal distribution for option pricing has been largely driven by empi... more The departure from the log-normal distribution for option pricing has been largely driven by empirical observations on skewness. Weibull distribution and the generalized beta distribution have been used recently to fit the risk-neutral density from option prices. Here we propose the use of the generalized gamma distribution for recovering the risk-neutral density. In terms of complexity, this distribution, having three parameters, falls between the Weibull and generalized beta. New op- tion pricing formulas for European calls and puts are derived under the generalized

Political risk is important to international investors because political events often have a dram... more Political risk is important to international investors because political events often have a dramatic effect on stock market performance. This paper seeks to analyze what the actual levels of political risk associated with Pakistan's stock markets have been in the past and what they are likely to be in the near future. Several Bayesian econometric models are applied to the series of counts of political events that influenced the evolution of stock market in Pakistan. The data is sparse and advanced fitting techniques such as Markov Chain Monte Carlo are employed to overcome difficulties related to obtaining inference within this context.

In this paper hypothesis tests are proposed for discrimination between the populations of two Alp... more In this paper hypothesis tests are proposed for discrimination between the populations of two Alpha distributions. This distribution is used for highly skewed data. The tests developed here are uniformly most powerful unbiased and can be used to test various general hypotheses related to this probability distribution which is less known by professional statisticians.

The generalized Rayleigh distribution is a two-parameter family of distributions having Rayleigh,... more The generalized Rayleigh distribution is a two-parameter family of distributions having Rayleigh, Maxwell and chi-square distributions as particular cases. We propose uniformly most powerful unbiased tests for discriminating between the parameters θ of two generalized Rayleigh distributions, conditional on the value of parameters k.

In this paper we model political risk for international capital budgeting as the value of a hypot... more In this paper we model political risk for international capital budgeting as the value of a hypothetical insurance policy that pays the holder any and all losses arising from political events. We address three important aspects of political risk that are widely acknowledged in the literature but either missing or incomplete in existing mathematical models: 1) loss causing political events

Handbook of Finance, 2008

... Risk Management in Freight Markets with Forwards and Options Contracts. Juby George 1 ,; Radu... more ... Risk Management in Freight Markets with Forwards and Options Contracts. Juby George 1 ,; Radu Tunaru Senior Lecturer PhD 2. Published Online: 15 SEP 2008. DOI: 10.1002/ 9780470404324.hof003012. Copyright © 2008 John Wiley & Sons, Inc. All rights reserved. Book ...

The main objective of this research is to identify probability dis- tributions that are capable t... more The main objective of this research is to identify probability dis- tributions that are capable to handle negative and positive skewness and can be applied to model the risk-neutral density function. The investigation will focus on comparing the flt of these distributions pro- vide in practice for fltting risk neutral densities with a view on pricing derivatives. The models are

The Journal of Derivatives, 2014

ABSTRACT This article describes a new methodology to compute the value at risk and the expected s... more ABSTRACT This article describes a new methodology to compute the value at risk and the expected shortfall using a power transformation technique. The methodology is an improvement of a recent method employing Johnson's system of distributions and is based on the idea of matching exactly the first four moments of the target portfolio distribution. The performance of this method is investigated in the context of jump-diffusion models with lognormal jumps. The authors show that it yields valid densities and quantile functions and that it is also superior to other techniques proposed in the literature, such as Cornish Fisher, Gram-Charlier, and Johnson distributions, in terms of relative error to the analytical benchmark. The power transformation is also applied to forecasting VaR for three equity indexes at different critical levels. The backtesting results support the validity of the newly proposed method.