• Selected
• Entire List

Abstract:
This exploratory research examines and models the financial distress prediction using neural network approach. The study is based on financial ratios. Nine different neural network models are constructed to test the predictive capability of the models by considering: (1) the impact of time varying information structure prior the distressed situation using first, independent annual financial ratios (four models)and second, different panel data sets (three models) and, (2) the influence of time varying probability estimates of financial distress in panel data sets (two models). Results support that it is not necessary to have complex architecture in neural models to predict firm's financial distress. Besides more the predictability horizon is shorter and the input information structure is most recent, more the predictive capability of the neural model is better.

financial distress, neural network, risk management

Abstract:
The aim of this paper is twofold; first we concentrate on the work of Vasicek (1977) and Cox, Ingersoll and Ross (1985). We examine and test empirically each model and discuss its performance in predicting the term structure of interest rates using a parametric estimating approach GMM (Generalized Moments Method). Second we estimate the term structure of interest rate dynamics using a nonparametric approach ANN (Artificial Neural Network). Two neural network models are performed. The first model uses spreads between interest rates of 10 different maturities as the only explanatory variable of interest rate changes. The second model introduces two factors, spreads and interest rates' levels. Using historical U.S. Treasury bill rates and Treasury bond yields, we compare the ability of each model to predict the term structure of interest rates. Data are daily and cover the period from 3 January 1995 to 29 December 2000. Results suggest that, neural network; Vasicek (1977) and Cox, Ingersoll and Ross (1985) models generate different yield curves. Neural network models outperform the parametric standard models. The most successful forecast is obtained with two factors neural network model.

Neural Network, Interest Rate Term Structure, Parametric models, GMM

Abstract:
This paper investigates whether international diversification and home bias inertia are substitutes or complements, using non-parametric stochastic dominance approach. More specifically we apply, on a data set consisting of daily closing prices of US stocks and index and Asian and Latin American stock market indices for the period from April 1st to October 29, 2004, Davidson and Duclos (2000) t-statistics and Barrett and Donald (2003), Monte Carlo and bootstrapped p-values to generate stochastic dominance relationships.

Empirical results show that international and domestic diversifications seem to be complementary strategies for American investors since no dominance relationships were revealed. However, substitutability can be shown according to risk aversion degree of investors. For low risk aversion coefficient, only domestic diversified portfolios first-second stochastically dominate global and international diversified portfolios in mean of 70 percent of cases. Inversely, for high risk aversion coefficient, global and international diversification strategies are preferred to efficient domestic diversification one. Besides, in mean of 60 percent of cases, global diversification can be substitute to international major and emerging diversification strategies for risk-adverse American investors.

Home bias and international diversification substitutability and complementary, Non-parametric stochastic dominance approach, Monte Carlo and Bootstrap p-values

Abstract:
The aim of this paper is to study the impact of Stock returns volatility of reference entities on credit default swap rates using a new dataset from the Japanese market. The majority of empirical research suggests the inadequacy of multinormal distribution and then the failure of methods based on correlation for measuring the structure of dependency. Using a copula approach, we can model the different relationships that can exist in different ranges of behavior. We study the bivariate distributions of credit default swap rates and the measure of stock return volatility estimated with GARCH (1,1) and focus on one parameter Archimedean copula. Starting from the empirical rank correlation statistics (Kendall's tau and Spearman's rho), we estimate the parameter values of each copula function presented in our study. Then, we choose the appropriate Archimedean copula that better fit to our data. We emphasize the finding that pairs with higher rating present a weaker dependence coefficient and then, the impact of stock return volatility on credit default swap rates is higher for the lowest rating class.

Copulas functions, credit default swap, volatility, bivariate distribution, Non-parametric estimation, Semi-parametric estimation

Abstract:
This paper deals with the impact of structure of dependency and the choice of procedures for rare-event simulation, on the pricing of multi-name credit derivatives such as basket credit default swap. A copula-based simulation procedure for pricing basket credit default swaps, under different structure of dependency, is presented here. The choice of copula and procedures for rare-event simulation govern the pricing of the basket credit default swap. Alternatives to the Gaussian copula are the Clayton copula and t-student copula under importance sampling procedures for simulation, which capture the dependence structure between the underlying variables at extreme values and certain values of the input random variables in a simulation, and have more impact, than others, on the parameter being estimated.

Abstract:
The aim of this paper is to estimate the fair spread of reconstituted basket credit default swap using Japanese market data. The value of these instruments depends on a number of factors including credit rating of the obligors in the basket, recovery rates, intensity of default, basket size, and the correlation of obligors in the basket. A fundamental part of the pricing framework is the estimation of the instantaneous default probabilities for each obligor. As default probabilities depend on the credit quality of the considered obligor, well-calibrated credit curves are a main ingredient for constructing default times. Similarly, the choice of copula for modeling the correlation of obligors in the basket and the choice of procedures for rare-event simulation govern the pricing of basket credit derivatives. The study has several practical implications that are of value to the financial hedgers and engineers, financial regulators, government regulators, central banks, and financial risk managers.

Abstract:
This paper examines and models the financial distress prediction using neural network approach. Nine different neural network models, considering various predicting time horizons and information structures, are considered. in order to test models' predictive capability we used a set of 15 financial ratios. Based on financial statements (balance-sheets, result accounts and cash flow statements) for 87 Tunisian firms from 1993 to 1996, results prove that more the predictability horizon is short and the input information structure recent, more and better is the predictive capability of the neural model. Short debt, capital structure and sales growth and liability ratios contribute meaningfully in discriminating and predicting the firm financial distress. the best model is based on the information structure giving the best predictive capability.

financial distress, neural networks, financial ratios

Abstract:
The aim of this paper is to explain empirically the determinants of credit default swap rates using a linear regression. We document that the majority of variables, detected from the credit risk pricing theories, explain more than 60% of the total level of credit default swap. These theoretical variables are credit rating, maturity, riskless interest rate, slope of the yield curve and volatility of equities. The estimated coefficients for the majority of these variables are consistent with theory and they are significant both statistically and economically. We conclude that credit rating is the most determinant of credit default swap rates.

credit derivatives, credit risk, rating, market variables