Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 7
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

On the use of multiple criteria distance indexes to find robust cash management policies

  • Salas-Molina, Francisco|||0000-0002-1168-7931
  • Pla Santamaría, David|||0000-0003-1563-3997
  • Rodriguez-Aguilar, Juan A.
[EN] Cash management decision-making can be handled from a multiobjective perspective by optimizing not only cost but also risk. Nevertheless, choosing the best policies under a changing context is by no means straightforward. To this end, we rely on compromise programming to incorporate robustness as an additional goal to cost and risk within a multiobjective framework. As a result, we propose to calculate robustness as a multiple criteria distance index that is able to identify the best compromise policies in terms of cost and risk. Such policies are also robust to cash flow regime changes. We show its utility by transforming the Miller and Orr s cash management model into its robust counterpart using real data from an industrial company., Ministerio de Economia y Competitividad [grant number Collectiveware TIN2015-66863-C2-1-R], [grant number 2014 SGR 118]. Work partially funded by projects Collectiveware TIN2015-66863-C2-1-R (MINECO/FEDER) and 2014 SGR 118.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Characterizing compromise solutions for investors with uncertain risk preferences

  • Salas-Molina, Francisco|||0000-0002-1168-7931
  • Pla Santamaría, David|||0000-0003-1563-3997
  • Rodriguez-Aguilar, Juan A.
[EN] The optimum portfolio selection for an investor with particular preferences was proven to lie on the normalized efficient frontier between two bounds defined by the Ballestero (1998) bounding theorem. A deeper understanding is possible if the decision-maker is provided with visual and quantitative techniques. Here, we derive useful insights as a way to support investor's decision-making through: (i) a new theorem to assess balance of solutions; (ii) a procedure and a new plot to deal with discrete efficient frontiers and uncertain risk preferences; and (iii) two quality metrics useful to predict long-run performance of investors., Work partially funded by projects Collectiveware TIN2015-66863-C2-1-R (MINECO/FEDER) and 2014 SGR 118
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

A multi-objective approach to the cash management problem

  • Salas-Molina, Francisco|||0000-0002-1168-7931
  • Pla Santamaría, David|||0000-0003-1563-3997
  • Rodriguez Aguilar, Juan Antonio
[EN] Cash management is concerned with optimizing costs of short-term cash policies of a company. Different optimization models have been proposed in the literature whose focus has been only placed on a single objective, namely, on minimizing costs. However, cash managers may also be interested in risk associated to cash policies. In this paper, we propose a multi-objective cash management model based on compromise programming that allows cash managers to select the best policies, in terms of cost and risk, according to their risk preferences. The model is illustrated through several examples using real data from an industrial company, alternative cost scenarios and two different measures of risk. As a result, we provide cash managers with a new tool to allow them deciding on the level of risk to take in daily decision-making., Work partially funded by projects Collectiveware TIN2015-66863-C2-1-R (MINECO/FEDER) and 2014 SGR 118.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Selecting cash management models from a multiobjective perspective

  • Salas-Molina, Francisco|||0000-0002-1168-7931
  • Díaz-García, Pablo|||0000-0002-7093-6061
  • Rodríguez-Aguilar, Juan A.
[EN] This paper addresses the problem of selecting cash management models under different operating conditions from a multiobjective perspective considering not only cost but also risk. A number of models have been proposed to optimize corporate cash management policies. The impact on model performance of different operating conditions becomes an important issue. Here, we provide a range of visual and quantitative tools imported from Receiver Operating Characteristic (ROC) analysis. More precisely, we show the utility of ROC analysis from a triple perspective as a tool for: (1) showing model performance; (2) choosingmodels; and (3) assessing the impact of operating conditions on model performance. We illustrate the selection of cash management models by means of a numerical example., Work partially funded by projects Collectiveware TIN2015-66863-C2-1-R (MINECO/FEDER) and 2014 SGR 118.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Empirical analysis of daily cash flow time series and its implications for forecasting

  • Salas-Molina, Francisco|||0000-0002-1168-7931
  • Rodriguez-Aguilar, Juan A.
  • Serrà,Joan
  • Guillen, Montserrat
  • Martin, Francisco J.
[EN] Usual assumptions on the statistical properties of daily net cash flows include normality, absence of correlation and stationarity. We provide a comprehensive study based on a real-world cash flow data set showing that: (i) the usual assumption of normality, absence of correlation and stationarity hardly appear; (ii) non-linearity is often relevant for forecasting; and (iii) typical data transformations have little impact on linearity and normality. This evidence may lead to consider a more data-driven approach such as time-series forecasting in an attempt to provide cash managers with expert systems in cash management., Work partially funded by projects Collectiveware TIN2015-66863-C2-1-R (MINECO/FEDER) and 2014 SGR 118.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Boundless multiobjective models for cash management

  • Salas-Molina, Francisco|||0000-0002-1168-7931
  • Pla Santamaría, David|||0000-0003-1563-3997
  • Rodriguez-Aguilar, Juan A
[EN] Cash management models are usually based on a set of bounds that complicate the selection of the optimal policies due to nonlinearity. We here propose to linearize cash management models to guarantee optimality through linear-quadratic multiobjective compromise programming models. We illustrate our approach through a reformulation of the suboptimal state-of-the-art Gormley-Meade¿s model to achieve optimality. Furthermore, we introduce a much simpler formulation that we call the boundless model that also provides optimal solutions without using bounds. Results from a sensitivity analysis using real data sets from 54 different companies show that our boundless model is highly robust to cash flow prediction errors., Generalitat de Catalunya [2014 SGR 118]; Ministerio de Economia y Competitividad [Collectiveware TIN2015-66863-C2-1-R].
Dipòsit Digital de Documents de la UAB
Publication Artículo científico (JournalArticle)

Coalition structure generation problems, optimization and parallelization of the IDP algorithm in multicore systems

  • Cruz Mencía, Francisco
  • Cerquides Bueno, Jesus|||0000-0002-3752-644X
  • Rodríguez-Aguilar, Juan A.
  • Espinosa, Antonio|||0000-0002-6460-3789
  • Moure, Juan C.|||0000-0001-6697-0331
  • Ramchurn, Sarvapali D.
  • Svensson, Kim
The coalition structure generation problem is well known in the area of multi-agent systems. Its goal is to establish coalitions between agents while maximizing the global welfare. Among the existing different algorithms designed to solve the coalition structure generation problem, DP and IDP are the ones with smaller temporal complexity. After analyzing the operation of the dynamic programming and improved dynamic programming algorithms, we have identified which are the most frequent operations and propose an optimized method. In addition, we study and implement a method for dividing the work into different threads. To describe incremental improvements of the algorithm design, we first compare performance of an improved single central processing unit core version where we obtain speedups ranging from 7 × to 11 × . Then, we describe the best resource use in a multi-thread optimized version where we obtain an additional 7.5 × speedup running in a 12-core machine.