Hall, A. D., Hwang, S. and Satchell, S., "Using Bayesian Variable Selection Methods to Choose Style Factors in Global Stock Return", March 2000.
Format: PDF, Size: 324 Kb

Abstract:

This paper applies Bayesian variable selection methods from the statistics literature to give guidance in the decision to include/omit factors in a global (linear factor) stock return model. Once one has accounted for country and sector, it is possible to see which style or styles best explains current asset returns. This study does not find compelling evidence for global styles, once country and sector have been accounted for.

Forbes, C. S. and Kofman, P., "Bayesian Target Zones", March 2000.

Abstract:

Several authors have postulated econometric models for exchange rates restricted to lie within known target zones. However, it is not uncommon to observe exchange rate data with known limits that are not fully 'credible'; that is, where some of the observations fall outside the stated range. An empirical model for exchange rates in a soft target zone where there is a controlled probability of the observed rates exceeding the stated limits is developed in this paper. A Bayesian approach is used to analyse the model, which is then demonstrated on Deutschemark-French franc and ECU-French franc exchange rate data.

Kofman, P. and Sharpe, I., "Imputation Methods for Incomplete Dependent Variables in Finance", March 2000.

Abstract:

Missing observations in dependent variables is a common feature of many financial applications. Standard ad hoc missing value imputation methods invariably fail to deliver efficient and unbiased parameter estimates. A number of recently developed classical and Bayesian iterative methods are evaluated for the treatment of missing dependent variables when the independent variables are completely observed. These methods are compared by simulation to commonly applied alternative missing data methodologies in the finance literature. The methods are then applied to a system of simultaneous equations modelling the maturity, secured status, and pricing of U.S. bank revolving loan contracts. Two of the four dependent varaibles in this application are characterised by severe missingness. The system of equations approach allows us to also exploit the additional information contained in the interdependencies among these features. The results indicate that proper treatement of missingness can be important for many financial applications.

Chiarella, C. and Kwon, O., "A Class of Heath-Jarrow-Morton Term Structure Models with Stochastic Volatility", March 2000.
Format: PDF, Size: 177 Kb

Abstract:

This paper considers a class of Heath-Jarrow-Morton term structure models with stochastic volatility. These models admit transformations to Markovian systems, and consequently lend themselves to well-established solution techniques for the bond and bond option prices. Solutions for certain special cases are obtained, and compared against their non-stochastic counterparts.

Chiarella, C. and He, X., "Heterogeneous Beliefs, Risk and Learning in a Simple Asset Pricing Model with a Market Maker", March 2000.
Format: PDF, Size: 6.1 Mb

Abstract:

This paper attempts to study the dynamics of a simple discounted present value asset price model where agents have different risk attitudes and follow different expectation formulation schemes for both first and second moments of the price distribution. Instead of using a Walrasian auctioneer scenario as the market clearing mechanism, a market maker scenario is used. In particular, the paper concentrates on models of fundamentalists and trend traders who follow least squares learning processes. An analysis is made of the effects of lag lengths on the stability of the fundamental equilibrium. Some necessary and/or sufficient conditions on the stability of the fundamental equilibrium associated with the speed of the adjustment of the market maker, different risk attitudes and different risk attitudes and different values of lags (used in the learning process) are established. The results lead to the following observations: (I) Compared with the findings with the Walrasian market clearing scenario in Brock and Hommes [8] and Chiarella and He [14], different price dynamics are obtained when the speed of the adjustment of the market maker increases and, in particular, when the contrarians are involved in the model; (ii) In contrast to homogeneous beliefs, where the larger is the lag length the more stable is thaat in general (for both the Walrasian and the market maker equilibrium) different lag lengths can complicate the price dynamics; (iii) In the model of trend followers versus contrarians, the stability of the fundamental equilibrium is determined by risk-adjusted aggregated extrapolation rates. This indicates that although every individual forecasting rule may lead to divergence from the equilibrium, these may "cancel out" in the aggregate and the actual dynamics with learning may thus be locally stable. On the other hand, only a small group of traders with expectation functions involving significant divergence can in fact destabilize the whole system.

Bhar, R., Chiarella, C., El-Hassan, N. and Zheng, X. "The Reduction of Forward Rate Dependent Volatility HJM Models to Markovian Form: Pricing European Bond Option", March 2000.
Format: PDF, Size: 132 Kb

Abstract:

We consider a single factor Heath-Jarrow-Morton model with a forward rate volatility function depending upon a function of time to maturity, the instantaneous spot rate of interest and a forward rate to a fixed maturity. With this specification the stochastic dynamics determining the prices of interest rate derivatives may be reduced to Markovian form. Furthermore, the evolution of the forward rate curve is completely determined by the two rates specified in the volatility function and it is thus possible to obtain a closed form expression for bond prices. The prices of bond options are determined by a partial differential equation involving two spatial variables. We discuss the evaluation of European bond options in this framework by use of the ADI method.

Chiarella, C. and He, X., "Stability of Competitive Equilibria with Heterogeneous Beliefs and Learning", March 2000.
Format: PDF, Size: 2.0 Mb

Abstract:

The paper studies a class of models in which agents' expectations influence the actual dynamics while the expectations themselves are the outcome of some learning processes. Situations of both homogeneous and heterogeneous beliefs are considered. In both cases agents update their expectations by general ah - and least-squares h-processes and the stability of the resulting dynamics in both cases is analysed. It is shown how the stability of the actual dynamics is affected by the heterogeneous expectations and different least-squares h-processes.

Lin, S. and Yang, J., "Examining Intraday Returns with Buy/Sell Information", March 2000.
Format: PDF, Size: 291 Kb

Abstract:

This paper examines high frequency stock returns with buy/sell signals. It demonstrates how such trading information could be utilized in a qualitative threshold framework to explain and predict the asymmetric behaviour of intrady stock returns. The study discovers that the buyer-dominating regime is consistently associated with negative returns, while the seller-dominating regime is consistently associated with positive returns. This is consistent with our suggestion of using the sign of the net buy/sell trading volume as the threshold indicator. Furthermore, the model renders better predicting power than that produced by a pure generalized autoregressive conditional heteroskedasticity model. Most interestingly, these reults are quite robust across all twelve actively traded stocks on the Australian Stock Exchange that we have examined, and hence provide strong support for the potential usefulness of buy/sell signals and the qualitative threshold model in analyzing the dynamics of high frequency financial asset returns.

Chiarella, C., Craddock, M. and El-Hassan, N., "The Calibration of Stock Option Pricing Models Using Inverse Problem Methodology", March 2000.
Format: PDF, Size: 218 Kb

Abstract:

We analyse the procedure for determining volatility presneted by Lagnado and Osher, and explain in some detail where the scheme comes from. We present an alternative scheme which avoids some of the technical complications arising in Lagnado and Osher's approach. An algorithm for solving the resulting equations is given, along with a selection of numerical examples.

Dunn, T., Schlögl, E. and Barton, G., "Simulated Swaption Delta-Hedging in the Lognormal Forward Libor Model", March 2000.
Format: PDF, Size: 380 Kb

Abstract:

Alternative approaches to hedging swaptions are explored and tested by simulation. Hedging methods implied by the Balck swaption formula are compared with a lognormal forward LIBOR model approach encompassing all the relevant forward rates. The simulation is undertaken within the LIBOR model framework for a range of swaptions and volatility structures. Despite incompatibilities with the model assumptions, the Black method performs equally well as the LIBOR method, yielding very similar distributions for the hedging profit and loss - even at high rehedging frequencies. This result demonstrates the robustness of the Black hedging technique and implies that - being simpler and generally better understood by financial practitioners - it would be the preferred method in practice.

Bhar, R., Chiarella, C. and Pham, T., "Modeling the Currency Forward Risk Premium: Theory and Evidence", April 2000.
Format: PDF, Size: 125 Kb

Abstract:

There is a huge literature on the existence of risk premia in the foreign exchange markets and its influence in explaining the divergence between the forward exchange rate and the subsequently realised spot exchange rate. In this paper, we seek to model directly the risk premium as a mean-reverting diffusion process. This is done by making use of the spot-forward price relationship and assuming a geometric Brownian process for the spot exchange rate. We are able to obtain a stochastic differential equation system for the spot exchange rate, the forward exchange rate and the risk premium which we estimate using Kalman filtering techniques. The model is then applied to the French Franc/USD and Japanese Yen/USD exchange rates from 1 January 1990 to 31 December 1998. For both currencies our main findings show (I) the persistence of substantial positive time variation in the forward risk premium and its alternating regimes; and (ii) the presence of a term structure of the forward risk premia.

Bhar, R. and Chiarella, C., "Infering Forward Looking Financial Market Risk Premia from Derivatives Prices", April 2000.
Format: PDF, Size: 84 Kb

Abstract:

This paper focuses on a topical and important area of theory and practice ie. The risk premium in financial markets. While there exists a vast amount of research into its behaviour, particularly in US markets, this is largely based on regression based techniques which do not capture well the dynamic and forward looking nature of the risk premium.

In this paper the variation of the unobserved risk premium is modelled by a system of stochastic differential equations connected by arbitrage arguments between the spot equity market, the index futures and options on index futures. Although various processes for the dynamic of the risk premium may be considered, we motivate and analyse a mean-reverting form. The diffusion part is specified such that the risk premium remains positive. Since the risk premium is not directly observable, information on it is extracted using an unobserved component state space formulation of the system and filtering methodology. As an initial application of the methodolgy, the results from daily Australian market data from the SFE over a period of twelve months are presented. The small sample properties of the parameter estimates are also examined by bootstrapping the state space system. It is well known in the filtering literature that the estimation of state space system by Kalman filter is sensitive to the specification of the quantities such as initial state variables and the prior covariance matrix. The paper also carries out approximate sensitivity analysis in this respect.

Chiarella, C. and Kwon, O., "A Complete Stochastic Volatility Model in the HJM Framework", July 2000.
Format: PDF, Size: 123 Kb

Abstract:

This paper considers a stochastic volatility version of the Heath, Jarrow and Morton (1992) term structure model. Market completeness is obtained by adapting the Hobson and Rogers (1998) complete stochastic volatility stock market model to the interest rate setting. Numerical simulation for a special case is used to compare the stochastic volatility model against the traditional Vasicek (1977) model.

Kuchler, U. and Platen, E., "Strong Discrete Time Approximation of Stochastic Differential Equations with Time Delay", September 2000.
Format: PDF, Size: 276 Kb

Abstract:

The paper introduces an approach for the derivation of discrete time approximations for solutions of stochastic differential equations with time delay. The suggested approximations converge in a strong sense. Furthermore, explicit solutions for linear stochastic delay equations are given. Keywords: stochastic differential equations with time delay; discrete time approximation; strong convergence; simulation

Platen, E., "Risk Premia and Financial Modelling Without Measure Transformation", September 2000.
Format: PDF, Size: 228 Kb

Abstract:

This paper describes a financial market modelling framework that exploits the notion of a deflator. The demonstrations of the deflator measured in units of primary assets form a minimal set of basic financial quantities that completely specify overall market dynamics. Risk premia of asset prices are obtained as a natural consequence of the approach. Contingent claim prices are computed under the real world measure both in the case of complete and incomplete markets.

Bohm, V. and Chiarella, C., "Mean Variance Preferences, Expectations Formation, and the Dynamics of Random Asset Prices", October 2000.
Format: PDF, Size: 627 Kb

Abstract:

This paper analyzes the dyanmics of a general explicit random price process of finitely many assets in an economy with overlapping generations of heterogeneous consumers forming optimal portfolios, extending the one dimensional investigation of Bohm, Deutscher and Wenzelburger (2000). Consumers maximize expected utility with respect to subjective transition probabilities defined by Markov kernels. Given a forecasting rule (predictor) and an exogeneous stochastic process of producer dividends, the dynamics of the economy is described as a random dynamical system in the sense of Arnold (1998). The paper investigates existence and stability of random fixed points (invariant measures) for mean-variance prefernces under various forecasting schemes, including unbiased predictions as well as OLS forecasting. Numerical simulations show the stability and the performance of the different predictors for linear mean-variance preferences. Alternative random dividend processes are provided.

Gerlach, R., Bird, R. and Hall, A. D., "A Bayesian Approach to Variable Selection in Logistic Regression with Application to Predicting Earnings Direction from Accounting Information", October 2000.
Format: PDF, Size: 235 Kb

Abstract:

This paper presents a Bayesian technique for the estimation of a logistic regression model including variable selection. The model is used, as in Ou and Penman (1989), to predict the direction of company earnings, one year ahead of time, from a large set of accounting variables from financial statements. We present a Markov chain Monte Carlo sampling scheme, that includes the variable slection technique of Smith and Kohn (1996) and the non-Gaussian estimation method of Mira and Tierney (1997), to estimate the model. The technique is applied to companies in the United States, United Kingdom and Australia. This extends the analysis of Ou and Penman (1989) who studied United States companies only. The results obtained comapre favourably to the technique used in Ou and Penamn (1989) for all three regions.