Forecasting correlated time series with exponential smoothing models
Abstract This paper presents the Bayesian analysis of a general multivariate exponential smoothing model that allows us to forecast time series jointly, subject to correlated random disturbances. The general multivariate model, which can be formulated as a seemingly unrelated regression model, includes the previously studied homogeneous multivariate Holt-Winters’ model as a special case when all of the univariate series share a common structure. MCMC simulation techniques are required in order to approach the non-analytically tractable posterior distribution of the model parameters. The predictive distribution is then estimated using Monte Carlo integration. A Bayesian model selection crite…
A fuzzy ranking strategy for portfolio selection applied to the Spanish stock market
In this paper we present a fuzzy ranking procedure for the portfolio selection problem. The uncertainty on the returns of each portfolio is approximated by means of a trapezoidal fuzzy number. The expected return and risk of the portfolio are then characteristics of that fuzzy number. A rank index that accounts for both expected return and risk is defined, allowing the decision-maker to compare different portfolios. The paper ends with an application of that fuzzy ranking strategy to the Spanish stock market.
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 investor's 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 …
Bayesian forecasting with the Holt–Winters model
Exponential smoothing methods are widely used as forecasting techniques in inventory systems and business planning, where reliable prediction intervals are also required for a large number of series. This paper describes a Bayesian forecasting approach based on the Holt–Winters model, which allows obtaining accurate prediction intervals. We show how to build them incorporating the uncertainty due to the smoothing unknowns using a linear heteroscedastic model. That linear formulation simplifies obtaining the posterior distribution on the unknowns; a random sample from such posterior, which is not analytical, is provided using an acceptance sampling procedure and a Monte Carlo approach gives …
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.
SIOPRED performance in a Forecasting Blind Competition
In this paper we present the results obtained by applying our automatic forecasting support system, named SIOPRED, over a data set of time series in a Forecasting Blind Competition. In order to apply our procedure for providing point forecasts it has been necessary to develop an interactive strategy for the choice of the suitable length of the seasonal cycle and the seasonality form for a generalized exponential smoothing method, which have been obtained using SIOPRED. For the choice of those essential characteristics of forecasting methods, also a certain multi-objective formulation which minimizes several measures of fitting is used. Once these specifications are established, the model pa…
Optimization under Uncertainty and Linear Semi-Infinite Programming: A Survey
This paper deals with the relationship between semi-infinite linear programming and decision making under uncertainty in imprecise environments. Actually, we have reviewed several set-inclusive constrained models and some fuzzy programming problems in order to see if they can be solved by means of a linear semi-infinite program. Finally, we present some numerical examples obtained by using a primal semi-infinite programming method.
A Forecasting Support System Based on Exponential Smoothing
This chapter presents a forecasting support system based on the exponential smoothing scheme to forecast time-series data. Exponential smoothing methods are simple to apply, which facilitates computation and considerably reduces data storage requirements. Consequently, they are widely used as forecasting techniques in inventory systems and business planning. After selecting the most adequate model to replicate patterns of the time series under study, the system provides accurate forecasts which can play decisive roles in organizational planning, budgeting and performance monitoring.
Convex semi-infinite games
This paper introduces a generalization of semi-infinite games. The pure strategies for player I involve choosing one function from an infinite family of convex functions, while the set of mixed strategies for player II is a closed convex setC inRn. The minimax theorem applies under a condition which limits the directions of recession ofC. Player II always has optimal strategies. These are shown to exist for player I also if a certain infinite system verifies the property of Farkas-Minkowski. The paper also studies certain conditions that guarantee the finiteness of the value of the game and the existence of optimal pure strategies for player I.
On the numerical treatment of linearly constrained semi-infinite optimization problems
Abstract We consider the application of two primal algorithms to solve linear semi-infinite programming problems depending on a real parameter. Combining a simplex-type strategy with a feasible-direction scheme we obtain a descent algorithm which enables us to manage the degeneracy of the extreme points efficiently. The second algorithm runs a feasible-direction method first and then switches to the purification procedure. The linear programming subproblems that yield the search direction involve only a small subset of the constraints. These subsets are updated at each iteration using a multi-local optimization algorithm. Numerical test examples, taken from the literature in order to compar…
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 metho…
Fuzzy Mathematical Programming for Portfolio Management
The classical portfolio selection problem was formulated by Markowitz in the 1950s as a quadratic programming problem in which the risk variance is minimized. Since then, many other models have been considered and their associated mathematical programming formulations can be viewed as dynamic, stochastic or static decision problems. In our opinion, the model formulation depends essentially on two factors: the data nature and the treatment given to the risk and return goals. In this communication, we consider several approaches to deal with the data uncertainty for different classical formulations of the portfolio problem. We make use of duality theory and fuzzy programming techniques to ana…
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 requir…
Forecasting time series with missing data using Holt's model
This paper deals with the prediction of time series with missing data using an alternative formulation for Holt's model with additive errors. This formulation simplifies both the calculus of maximum likelihood estimators of all the unknowns in the model and the calculus of point forecasts. In the presence of missing data, the EM algorithm is used to obtain maximum likelihood estimates and point forecasts. Based on this application we propose a leave-one-out algorithm for the data transformation selection problem which allows us to analyse Holt's model with multiplicative errors. Some numerical results show the performance of these procedures for obtaining robust forecasts.
Viability of infeasible portfolio selection problems: A fuzzy approach
Abstract This paper deals with fuzzy optimization schemes for managing a portfolio in the framework of risk–return trade-off. Different models coexist to select the best portfolio according to their respective objective functions and many of them are linearly constrained. We are concerned with the infeasible instances of such models. This infeasibility, usually provoked by the conflict between the desired return and the diversification requirements proposed by the investor, can be satisfactorily avoided by using fuzzy linear programming techniques. We propose an algorithm to repair infeasibility and we illustrate its performance on a numerical example.
Optimality conditions for nondifferentiable convex semi-infinite programming
This paper gives characterizations of optimal solutions to the nondifferentiable convex semi-infinite programming problem, which involve the notion of Lagrangian saddlepoint. With the aim of giving the necessary conditions for optimality, local and global constraint qualifications are established. These constraint qualifications are based on the property of Farkas-Minkowski, which plays an important role in relation to certain systems obtained by linearizing the feasible set. It is proved that Slater's qualification implies those qualifications.
New descent rules for solving the linear semi-infinite programming problem
The algorithm described in this paper approaches the optimal solution of a continuous semi-infinite linear programming problem through a sequence of basic feasible solutions. The descent rules that we present for the improvement step are quite different when one deals with non-degenerate or degenerate extreme points. For the non-degenerate case we use a simplex-type approach, and for the other case a search direction scheme is applied. Some numerical examples illustrating the method are given.
Introducing a Fuzzy-Pattern Operator in Fuzzy Time Series
In this paper we introduce a fuzzy pattern operator and propose a new weighting fuzzy time series strategy for generating accurate ex-post forecasts. A decision support system is built for managing the weights of the information provided by the historical data, under a fuzzy time series framework. Our procedure analyzes the historical performance of the time series using different experiments, and it classifies the characteristics of the series through a fuzzy operator, providing a trapezoidal fuzzy number as one-step ahead forecast. We also present some numerical results related to the predictive performance of our procedure with time series of financial data sets.
A spreadsheet modeling approach to the Holt–Winters optimal forecasting
Abstract The objective of this paper is to determine the optimal forecasting for the Holt–Winters exponential smoothing model using spreadsheet modeling. This forecasting procedure is especially useful for short-term forecasts for series of sales data or levels of demand for goods. The non-linear programming problem associated with this forecasting model is formulated and a spreadsheet model is used to solve the problem of optimization efficiently. Also, a spreadsheet makes it possible to work in parallel with various objective functions (measures of forecast errors) and different procedures for calculating the initial values of the components of the model. Using a scenario analysis, the se…
Solving a class of fuzzy linear programs by using semi-infinite programming techniques
This paper deals with a class of Fuzzy Linear Programming problems characterized by the fact that the coefficients in the constraints are modeled as LR-fuzzy numbers with different shapes. Solving such problems is usually more complicated than finding a solution when all the fuzzy coefficients have the same shape. We propose a primal semi-infinite algorithm as a valuable tool for solving this class of Fuzzy Linear programs and, we illustrate it by means of several examples.
Multivariate exponential smoothing: A Bayesian forecast approach based on simulation
This paper deals with the prediction of time series with correlated errors at each time point using a Bayesian forecast approach based on the multivariate Holt-Winters model. Assuming that each of the univariate time series comes from the univariate Holt-Winters model, all of them sharing a common structure, the multivariate Holt-Winters model can be formulated as a traditional multivariate regression model. This formulation facilitates obtaining the posterior distribution of the model parameters, which is not analytically tractable: simulation is needed. An acceptance sampling procedure is used in order to obtain a sample from this posterior distribution. Using Monte Carlo integration the …
A purification algorithm for semi-infinite programming
Abstract In this paper we present a purification algorithm for semi-infinite linear programming. Starting with a feasible point, the algorithm either finds an improved extreme point or concludes with the unboundedness of the problem. The method is based on the solution of a sequence of linear programming problems. The study of some recession conditions has allowed us to establish a weak assumption for the finite convergence of this algorithm. Numerical results illustrating the method are given.
Fuzzy Portfolio Selection Models: A Numerical Study
In this chapter we analyze the numerical performance of some possibilistic models for selecting portfolios in the framework of risk-return trade-off. Portfolio optimization deals with the problem of how to allocate wealth among several assets, taking into account the uncertainty involved in the behavior of the financial markets. Different approaches for quantifying the uncertainty of the future return on the investment are considered: either assuming that the return on every individual asset is modeled as a fuzzy number or directly measuring the uncertainty associated with the return on a given portfolio. Conflicting goals representing the uncertain return on and risk of a fuzzy portfolio a…
A fuzzy decision support tool for demand forecasting
In this paper we present a decision support forecasting system to work with univariate time series based on the generalized exponential smoothing (Holt-Winters) approach. It is conceived as an integrated tool which has been implemented in Visual Basic. For improving the accuracy of the automatic forecasting it uses an optimization-based scheme which unifies the stages of estimation of the parameters 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. A set of real data is analyzed to show the performance of our forecasting too…
Optimizing the level of service quality of a bike-sharing system
Public bike-sharing programs have been deployed in hundreds of cities worldwide, improving mobility in a socially equitable and environmentally sustainable way. However, the quality of the service is drastically affected by imbalances in the distribution of bicycles among stations. We address this problem in two stages. First, we estimate the unsatisfied demand (lack of free lockers or lack of bicycles) at each station for a given time period in the future and for each possible number of bicycles at the beginning of the period. In a second stage, we use these estimates to guide our redistribution algorithms. Computational results using real data from the bike-sharing system in Palma de Mall…
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…
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 dist…
Measuring Uncertainty in the Portfolio Selection Problem
In this paper, we propose a new index for ranking portfolios based on the credibility expected return and loss on their investment. We assume that the return on a given portfolio is modeled as a trapezoidal fuzzy variable, whose credibility distribution is built using the data set of its historical returns. The credibilistic loss on the investment for a given portfolio is measured by means of a suitable loss function. In order to take risk-adverse investor attitudes into account, we analyze the performance of some credibility measures related to loss and risk on the investment for a given portfolio and their relationship with similar possibility measures. A numerical example is presented sh…
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 o…
Forecasting portfolio returns using weighted fuzzy time series methods
We propose using weighted fuzzy time series (FTS) methods to forecast the future performance of returns on portfolios. We model the uncertain parameters of the fuzzy portfolio selection models using a possibilistic interval-valued mean approach, and approximate the uncertain future return on a given portfolio by means of a trapezoidal fuzzy number. Introducing some modifications into the classical models of fuzzy time series, based on weighted operators, enables us to generate trapezoidal numbers as forecasts of the future performance of the portfolio returns. This fuzzy forecast makes it possible to approximate both the expected return and the risk of the investment through the value and a…
A new approach to portfolio selection based on forecasting
In this paper we analyze the portfolio selection problem from a novel perspective based on the analysis and prediction of the time series corresponding to the portfolio’s value. Namely, we define the value of a particular portfolio at the time of its acquisition. Using the time series of historical prices of the different financial assets, we calculate backward the value that said portfolio would have had in past time periods. A damped trend model is then used to analyze this time series and to predict the future values of the portfolio, providing estimates of the mean and variance for different forecasting horizons. These measures are used to formulate the portfolio selection problem, whic…
Improving stock index forecasts by using a new weighted fuzzy-trend time series method
Define a new technical indicator for measuring the trend of the fuzzy time series.Introduce a new weighted fuzzy-trend time series method to forecast stock indices.Compare ex-post performances of weighted FTS methods using stock market indices.Assess statistical significance of ex-post forecast accuracy for weighted FTS methods. We propose using new weighted operators in fuzzy time series to forecast the future performance of stock market indices. Based on the chronological sequence of weights associated with the original fuzzy logical relationships, we define both chronological-order and trend-order weights, and incorporate our proposals for the ex-post forecast into the classical modeling…
Bayesian forecasting of demand time-series data with zero values
This paper describes the development of a Bayesian procedure to analyse and forecast positive demand time-series data with a proportion of zero values and a high level of variability for the non-zero data. The resulting forecasts play decisive roles in organisational planning, budgeting, and performance monitoring. Exponential smoothing methods are widely used as forecasting techniques in industry and business. However, they can be unsuitable for the analysis of non-negative demand time-series data with the aforementioned features. In this paper, an unconstrained latent demand underlying the observed demand is introduced into the linear heteroscedastic model associated with the Holt-Winters…
Portfolios with fuzzy returns: Selection strategies based on semi-infinite programming
AbstractThis paper provides new models for portfolio selection in which the returns on securities are considered fuzzy numbers rather than random variables. The investor's problem is to find the portfolio that minimizes the risk of achieving a return that is not less than the return of a riskless asset. The corresponding optimal portfolio is derived using semi-infinite programming in a soft framework. The return on each asset and their membership functions are described using historical data. The investment risk is approximated by mean intervals which evaluate the downside risk for a given fuzzy portfolio. This approach is illustrated with a numerical example.
A multi-local optimization algorithm
The development of efficient algorithms that provide all the local minima of a function is crucial to solve certain subproblems in many optimization methods. A “multi-local” optimization procedure using inexact line searches is presented, and numerical experiments are also reported. An application of the method to a semi-infinite programming procedure is included.