A Hybrid Model for Estimation of Volatility of Call Option Price Using Particle Filter Sunil Kumar Dhal 1 , Prof.( Dr.) Srinivash Prasad 2 , Prof. (Dr.) Manojranjan Nayak 3 1 Associate Professor, Regional College of Management Chandrasekharpur, Bhubaneswar, Odisa, India 2 Dean( Academic),Gandhi Institute of Technological and Advancement Gohirapara, Bhubaneswar, Odisa, India 3 Chairman, Institute of Technical Education and Research Gangapatana, Bhubaneswar, Odisa, India Abstract In the recent years, the distribution of possible future losses for portfolios, such as bonds or loans, exhibits strongly asymmetric behavior. In this paper, we have analyzed the effective portfolio risk management through a computational state space model by using particle filter through sequential estimation of volatility. The computational model comprises with Extended weight Moving Average Model and Black Scholes-Option Pricing model as well as GARCH deterministic volatility model. The outcome of the model establishes the effectiveness of particle filter for estimating volatility of call option prices for future portfolio returns and it can able to predict the investor’s financial risk and measures in a significant manner. Keywords: Portfolio, financial risk, volatility, particle filter, call option, put option. 1. INTRODUCTION The volatility of a stock is defined as the measure of variation of price of a financial instrument over a time period . When the time period of interest is one year, then the volatility is an annual volatility year  and when the time period of interest is one day, then the volatility is a daily volatility Whatsoever, annual volatility is frequently estimated by first estimation daily volatility using daily log stock returns data. The three main purposes of Estimating volatility are for risk management, for asset allocation, and for taking bets on future volatility. A large part of risk management is measuring the potential future losses of a portfolio of assets, and in order to measure these potential losses, estimates must be made of future volatilities and correlations The Black-Scholes partial deferential equation and ultimately solve the equation for a European call option In the BSOPM (Black Scholes-Option Pricing model) framework, the annual volatility is taken as constant. It employs a common method which simply calculates the sample standard deviation of the daily log returns of the stock over the past N days by using the equations as below: 2 1 1 ( ) 1 n t i i S R R n      (1) Where the average value of the stock return is given as    n i i R n R 1 1 Where n is the number of stock return and ) / ln( 1   i i i S S R (2) S i gives us an estimate of daily volatility. t  . Since year  is an annual quantity; we have to scale or estimate St. year  which is estimated by t year s 1/TD   (3) Where TD is the annual number of Trading Days (TD) To simulate return values for testing the methods, we will use the stochastic differential equation that corresponds to geometric Brownian motion, dX dt S dS       where ) (t S is the stock price at time t,  is a measure of the average rate of growth of the asset price, dt is the change in time,  is the volatility, and dX is known as a Wiener process because it is a random normal variable with a mean of zero and a standard deviation of dt For the numerical simulation the initial asset price was set equal to p, Where p is a numerical value. In terms of the model 0 S = p, and St is the closing price for day t. IJCSI International Journal of Computer Science Issues, Vol. 9, Issue 4, No 1, July 2012 ISSN (Online): 1694-0814 www.IJCSI.org 459 Copyright (c) 2012 International Journal of Computer Science Issues. All Rights Reserved.