Alessandra Amendola

A GMM procedure for combining volatility forecasts

A novel approach to the combination of volatility forecasts is discussed. The proposed procedure makes use of the generalized method of moments (GMM) for estimating the combination weights. The asymptotic properties of the GMM estimator are derived while its finite sample properties are assessed by means of a simulation study. The results of an application to a time series of daily returns on the S&P500 are presented.

Combination of multivariate volatility forecasts

This paper proposes a novel approach to the combination of conditional covariance matrix forecasts based on the use of the Generalized Method of Moments (GMM). It is shown how the procedure can be generalized to deal with large dimensional systems by means of a two-step strategy. The finite sample properties of the GMM estimator of the combination weights are investigated by Monte Carlo simulations. Finally, in order to give an appraisal of the economic implications of the combined volatility predictor, the results of an application to tactical asset allocation are presented.

Forecasting corporate bankruptcy: empirical evidence on Italian data

Purpose – The aim of the paper is to investigate several aspects of bankruptcy prediction within both theoretical and empirical frameworks. In particular, it has focused on the comparison of different techniques used to forecast failure through a balanced sample of companies within a geographical area (the Campania region) located in the south of Italy.Methodology – Business failure has been one of the most investigated topics within corporate finance and the empirical approach to bankruptcy prediction has recently gained further attention from financial institutions. The aim of corporate failure prediction is to have a methodological approach which discriminates firms with a high probability of future failure from those which are considered to be healthy. Starting from the seminal paper of Altman (1968), many other significant contributions have been subsequently made to this field (Ravi Kumar and Ravi, 2007). This papers approach is to compare different statistical techniques based on the analysis of f...

A NON LINEAR TIME SERIES APPROACH TO MODELLING ASYMMETRY IN STOCK MARKET INDEXES

In this paper we analyse the performances of a novel approach to modelling non-linear conditionally heteroscedastic time series characterised by asymmetries in both the conditional mean and variance. This is based on the combination of a TAR model for the conditional mean with a Constrained Changing Parameters Volatility (CPV-C) model for the conditional variance. Empirical results are given for the daily returns of the S&P 500, NASDAQ composite and FTSE 100 stock market indexes.

Temporal Aggregation and Closure of VARMA Models: Some New Results

In this paper we examine the effects of temporal aggregation on Vector AutoRegressive Moving Average (VARMA) models. It has relevant implications both in theoretical and empirical domain. Among them we focus the attention on the main consequences of the aggregation (obtained from point in time sampling) on the model identification. Further, under well defined conditions on the model parameters, we explore the closure of the VARMA class (with respect to the temporal aggregation) through theoretical results discussed in proper examples.

Statistical Properties of Threshold Models

This article focuses the attention on the Self Exciting Threshold Autoregressive Moving Average model (SETARMA) proposed in Tong (1983). The stochastic structure of the model is discussed and different specifications are presented. Starting from one of them, we give sufficient conditions for the weak stationarity of the model that are discussed and critically compared to other results given in literature. In particular, after showing that the SETARMA model belongs to the class of the Random Coefficients Autoregressive models, widely discussed in Nicholls and Quinn (1982), we give some issues on the weak stationarity of its stochastic structure that are more general than those given in the existing literature and appear not affected by the moving average component.

Editorial: Special Issue on Statistical and Computational Methods in Finance

In recent decades major developments in computational methods allowed revolutionary changes to take place in the statistical and econometric analysis of financial processes. Evaluating various forecasts and policy scenarios with their implied risk using advanced computational techniques and modern financial models is becoming more and more standard practice. The contents of this special issue reflect the growing interest in this area of research. The journal Computational Statistics and Data Analysis has regular issues on computational and financial econometrics, and statistical methods in finance. Of particular interest are papers in important areas of econometric and financial applications where both computational techniques and numerical methods have a major impact (Amendola et al., 2006; Belsley et al., 2007; Geweke et al., 2007; Gilli and Winker, 2007; Pollock and Proietti, 2007). The goal is to provide sources of information about the most recent developments in computational econometrics that are currently scattered throughout publications in specialized areas. This special issue comprises 18 articles covering a wide range of topics such as dynamic evolution of the volatility of financial returns, model-free measurements of volatility, combination of volatility forecasts, and a transmission mechanism of volatility between markets and operational risk management. Several methodological approaches are proposed based on estimation methods such as MLE, GMM and Bayes and also based on techniques such as copulas, wavelets and geostatistical procedures. Ruiz and Veiga (this issue) focus on leverage effects and long-memory in volatility. They present a new model: the asymmetric long-memory stochastic volatility (A-LMSV) model to cope with leverage effects. Statistical properties of the new model are derived and compared with the properties of the FIEGARCH model. The results are illustrated by fitting both models so as to represent the dynamic evolution of volatilities of daily returns of the S&P500 and DAX indexes. The paper by Creal (this issue) compares alternative filtering and smoothing algorithms for estimating stochastic volatility (SV) models when realized volatility is used as an observable measure of the unobserved true volatility. The author examines how well the particle filter compares with the Kalman filter in estimating the integrated variance under a number of different specifications of the model. Lindström et al. (this issue) introduce a framework based on the state-space formulation of the option valuation model. The performance and computational efficiency of standard and iterated extended Kalman filters are investigated. The tracking time-varying parameters and latent processes such as SV processes have also been studied through a simulation. Omori and Watanabe (this issue) propose an efficient Bayesian method using Monte Carlo Markov Chain (MCMC) for the estimation of asymmetric SV models. They extend their previous results to develop a block sampler method that can take into account asymmetric effects in the returns. The paper by Strickland et al. (this issue) examines the effects of parameterization on the simulation efficiency of MCMC algorithms for non-Gaussian state-space models. Specifically, the authors consider the stochastic conditional duration (SCD) and the SV models. They investigate four alternative parameterizations: centred, non-centred in location, non-centred in scale and non-centred in both location and scale. The relations among the simulation efficiency of the MCMC sampler, the magnitudes of the population parameters and the particular parameterizations of the state-space model are examined.

Evaluation of volatility predictions in a VaR framework

The evaluation of volatility forecasts is not straightforward and some issues can arise. A standard approach relies on statistical loss functions. Another approach bases the evaluation of the volatility predictions on utility functions or Value at Risk (VaR) measures. This work aims to combine the two approaches, using the VaR measures within the loss functions. By means of this method, the VaR measures obtained from a set of competing models are plugged into two loss functions, the magnitude loss function and a proposed new one. This latter loss function more heavily penalizes the models with a number of VaR violations greater than the expected one. The loss function values are evaluated against a benchmark obtained from the inclusion of a consistent estimate of the VaR measures in the loss function. In order to investigate the performance of the proposed method and the new loss function, a Monte Carlo experiment and an empirical analysis of a stock listed on the New York Stock Exchange are provided. The proposed strategy helps with the selection of a superior model, in terms of forecast accuracy, when the cited approaches do not clearly and uniquely identify it. Moreover, the new asymmetric loss function allows a greater discrimination with regard to models, helping to find the best volatility model.

Least Squares Predictors for Threshold Models: Properties and Forecast Evaluation

The forecasts generation from models that belong to the threshold class is discussed. The main problems that arise when forecasts have to be computed from these models are presented and, in particular, least squares, plug-in and combined predictors are pointed out. The performance of the proposed predictors are investigated using simulated and empirical examples that give evidence in favor of the forecasts combination.

The Autocorrelation Functions in SETARMA Models

The dependence structure of a family of self exciting threshold autoregressive moving average (SETARMA) models, is investigated. An alternative representation for this class of models is proposed and the exact autocorrelation function is derived in the case of two regimes. Some practical implications of the theoretical results are analysed and discussed via several examples of SETARMA structures of fixed orders