José D. Bermúdez

Fuzzy portfolio optimization under downside risk measures

This paper presents two fuzzy portfolio selection models where the objective is to minimize the downside risk constrained by a given expected return. We assume that the rates of returns on securities are approximated as LR-fuzzy numbers of the same shape, and that the expected return and risk are evaluated by interval-valued means. We establish the relationship between those mean-interval definitions for a given fuzzy portfolio by using suitable ordering relations. Finally, we formulate the portfolio selection problem as a linear program when the returns on the assets are of trapezoidal form.

Poor Survival for Infants With MYCN-Amplified Metastatic Neuroblastoma Despite Intensified Treatment: The International Society of Paediatric Oncology European Neuroblastoma Experience

PURPOSE To report the results of a prospective, nonrandomized European study on infants with neuroblastoma and MYCN gene amplification. PATIENTS AND METHODS Infants with neuroblastoma (stage 2, 3, 4, and 4s) and MYCN gene amplification who were diagnosed between 1999 and 2004 were eligible for enrollment onto the study. After diagnosis, staging, and mandatory biologic studies, induction chemotherapy (IC) with conventional drugs was administered, followed by delayed surgery, megatherapy (busulfan-melphalan as a conditioning regimen), and local radiotherapy. RESULTS Of the 46 infants enrolled onto the study, 35 infants were eligible; of these 35 infants, 97% had metastatic spread (24 infants had stage 4, and 10 infants had stage 4s). Two-year overall survival (OS) was 30% (SE, 0.08), with median survival time of 12 months, and 23 deaths due to disease. Two-year, event-free survival (EFS) was 29% (SE, 0.07). The treatment was well tolerated with no deaths as a result of toxicity or severe toxicity. Despite protocol adherence, 30% of the patients who were assessable for response to IC experienced disease progression or did not respond. Stage and high lactate dehydrogenase reached significance in the univariate analysis (P = .028 and .039, respectively for OS; and P = .05 and .031 respectively, for EFS). Ten of 16 patients who received megatherapy are still alive. CONCLUSION Although treatment was well tolerated, survival was poor and our IC failed to achieve a satisfactory response in 30% of our patients. New therapeutic approaches and more intense world-wide collaboration are needed to achieve a cure in this population.

A multi-objective genetic algorithm for cardinality constrained fuzzy portfolio selection

This paper presents a new procedure that extends genetic algorithms from their traditional domain of optimization to fuzzy ranking strategy for selecting efficient portfolios of restricted cardinality. The uncertainty of the returns on a given portfolio is modeled using fuzzy quantities and a downside risk function is used to describe the investors aversion to risk. The fitness functions are based both on the value and the ambiguity of the trapezoidal fuzzy number which represents the uncertainty on the return. The soft-computing approach allows us to consider uncertainty and vagueness in databases and also to incorporate subjective characteristics into the portfolio selection problem. We use a data set from the Spanish stock market to illustrate the performance of our approach to the portfolio selection problem.

Hypermethylation of apoptotic genes as independent prognostic factor in neuroblastoma disease

Neuroblastoma (NB) is an embryonal tumour of neuroectodermal cells, and its prognosis is based on patient age at diagnosis, tumour stage and MYCN amplification, but it can also be classified according to their degree of methylation. Considering that epigenetic aberrations could influence patient survival, we studied the methylation status of a series of 17 genes functionally involved in different cellular pathways in patients with NB and their impact on survival. We studied 82 primary NB tumours and we used methylation‐specific‐PCR to perform the epigenetic analysis. We evaluated the putative association among the evidence of hypermethylation with the most important NB prognostic factors, as well as to determine the relationship among methylation, clinical classification and survival. CASP8 hypermethylation showed association with relapse susceptibility and, TMS1 and APAF1 hypermethylation are associated with bad prognosis and showed high influence on NB overall survival. Hypermethylation of apoptotic genes has been identified as a good candidate of prognostic factor. We propose the simultaneous analysis of hypermethylation of APAF1, TMS1 and CASP8 apoptotic genes on primary NB tumour as a good prognostic factor of disease progression. Mol. Carcinog.

Evolutionary multi-objective optimization algorithms for fuzzy portfolio selection

Graphical abstractDisplay Omitted HighlightsWe consider a constrained three-objective optimization portfolio selection problem.We solve the problem by means of evolutionary multi-objective optimization.New mutation, crossover and reparation operators are designed for this problem.They are tested in several algorithms for a data set from the Spanish stock market.Results for two performance metrics reveal the effectiveness of the new operators. In this paper, we consider a recently proposed model for portfolio selection, called Mean-Downside Risk-Skewness (MDRS) model. This modelling approach takes into account both the multidimensional nature of the portfolio selection problem and the requirements imposed by the investor. Concretely, it optimizes the expected return, the downside-risk and the skewness of a given portfolio, taking into account budget, bound and cardinality constraints. The quantification of the uncertain future return on a given portfolio is approximated by means of LR-fuzzy numbers, while the moments of its return are evaluated using possibility theory. The main purpose of this paper is to solve the MDRS portfolio selection model as a whole constrained three-objective optimization problem, what has not been done before, in order to analyse the efficient portfolios which optimize the three criteria simultaneously. For this aim, we propose new mutation, crossover and reparation operators for evolutionary multi-objective optimization, which have been specially designed for generating feasible solutions of the cardinality constrained MDRS problem. We incorporate the operators suggested into the evolutionary algorithms NSGAII, MOEA/D and GWASF-GA and we analyse their performances for a data set from the Spanish stock market. The potential of our operators is shown in comparison to other commonly used genetic operators and some conclusions are highlighted from the analysis of the trade-offs among the three criteria.

Holt–Winters Forecasting: An Alternative Formulation Applied to UK Air Passenger Data

Abstract This paper provides a formulation for the additive Holt–Winters forecasting procedure that simplifies both obtaining maximum likelihood estimates of all unknowns, smoothing parameters and initial conditions, and the computation of point forecasts and reliable predictive intervals. The stochastic component of the model is introduced by means of additive, uncorrelated, homoscedastic and Normal errors, and then the joint distribution of the data vector, a multivariate Normal distribution, is obtained. In the case where a data transformation was used to improve the fit of the model, cumulative forecasts are obtained here using a Monte-Carlo approximation. This paper describes the method by applying it to the series of monthly total UK air passengers collected by the Civil Aviation Authority, a long time series from 1949 to the present day, and compares the resulting forecasts with those obtained in previous studies.

A decision support system methodology for forecasting of time series based on soft computing

Exponential procedures are widely used as forecasting techniques for inventory control and business planning. A number of modifications to the generalized exponential smoothing (Holt-Winters) approach to forecasting univariate time series is presented, which have been adapted into a tool for decision support systems. This methodology unifies the phases of estimation and model selection into just one optimization framework which permits the identification of robust solutions. This procedure may provide forecasts from different versions of exponential smoothing by fitting the updated formulas of Holt-Winters and selects the best method using a fuzzy multicriteria approach. The elements of the set of local minima of the non-linear programming problems allow us to build the membership functions of the conflicting objectives. It is compared to other forecasting methods on the 111 series from the M-competition.

A Possibilistic Mean-Downside Risk-Skewness Model for Efficient Portfolio Selection

This paper presents a new possibilistic model for the portfolio selection problem. The uncertainty of future returns on a given portfolio is modeled using LR-fuzzy numbers. Some possibilistic moments are considered to measure the risk of and return on the investment. Since the joint possibility distribution of the returns on the assets is unknown, we consider the returns on a given portfolio as the historical dataset instead of considering the individual returns on the assets as the dataset. We introduce a coefficient of possibilistic skewness in order to incorporate a measurement of the asymmetry of the fuzzy return on a given portfolio. We solve the multi-objective optimization problems that are associated with the possibilistic mean-downside risk-skewness model by using an evolutionary procedure to generate efficient portfolios. The procedure provides different patterns of investment, whose portfolios meet the explicit restrictions imposed by the investor. Thus, from among the points in the efficient frontier, the investor may select a portfolio that optimizes an economically meaningful objective function. The performance of this approach is tested using a dataset of assets from the Spanish stock market.

Improving demand forecasting accuracy using nonlinear programming software

We address the problem of forecasting real time series with a proportion of zero values and a great variability among the nonzero values. In order to calculate forecasts for a time series, the model coefficients must be estimated. The appropriate choice of values for the smoothing parameters in exponential smoothing methods relies on the minimization of the fitting errors of historical data. We adapt the generalized Holt–Winters formulation so that it can consider the starting values of the local components of level, trend and seasonality as decision variables of the nonlinear programming problem associated with this forecasting procedure. A spreadsheet model is used to solve the problems of optimization efficiently. We show that our approach produces accurate forecasts with little data per product.

Portfolio optimization using a credibility mean-absolute semi-deviation model

We present a cardinality constrained credibility mean-absolute semi-deviation model.We prove relationships for possibility and credibility moments for LR-fuzzy variables.The return on a given portfolio is modeled by means of LR-type fuzzy variables.We solve the portfolio selection problem using an evolutionary procedure with a DSS.We select best portfolio from Pareto-front with a ranking strategy based on Fuzzy VaR. We introduce a cardinality constrained multi-objective optimization problem for generating efficient portfolios within a fuzzy mean-absolute deviation framework. We assume that the return on a given portfolio is modeled by means of LR-type fuzzy variables, whose credibility distributions collect the contemporary relationships among the returns on individual assets. To consider credibility measures of risk and return on a given portfolio enables us to work with its Fuzzy Value-at-Risk. The relationship between credibility expected values for LR-type fuzzy variables and possibilistic moments for LR-fuzzy numbers having the same membership function are analyzed. We apply a heuristic approach to approximate the cardinality constrained efficient frontier of the portfolio selection problem considering the below-mean absolute semi-deviation as a measure of risk. We also explore the impact of adding a Fuzzy Value-at-Risk measure that supports the investors choices. A computational study of our multi-objective evolutionary approach and the performance of the credibility model are presented with a data set collected from the Spanish stock market.