https://www.um.edu.mt/library/oar/handle/123456789/69722 Mon, 06 Apr 2026 22:01:52 GMT 2026-04-06T22:01:52Z https://www.um.edu.mt/library/oar/handle/123456789/70847 Title: How to classify a government can a perceptron do it? Authors: Caleiro, António Abstract: The electoral cycle literature has developed in two clearly distinct phases. The first one considered the existence of non-rational (naive) voters whereas the second one considered fully rational voters. It is our view that an intermediate approach is more appropriate, i.e. one that considers learning voters, which are boundedly rational. In this sense, one may consider perceptrons as learning mechanisms used by voters to perform a classification of the incumbent in order to distinguish opportunistic (electorally motivated) from benevolent (non-electorally motivated) behaviour of the government. The paper explores precisely the problem of how to classify a government showing in which, if so, circumstances a perceptron can resolve that problem. This is done by considering a model recently considered in the literature, i.e. one allowing for output persistence, which is a feature of aggregate supply that, indeed, may turn impossible to correctly classify the government. Tue, 01 Jan 2013 00:00:00 GMT https://www.um.edu.mt/library/oar/handle/123456789/70847 2013-01-01T00:00:00Z https://www.um.edu.mt/library/oar/handle/123456789/70846 Title: Combining time series analysis and multi criteria decision making techniques for forecasting financial performance of banks in Turkey Authors: Önder, Emrah; Hepşen, Ali Abstract: Forecasting plays a major role in financial planning and it is an essential analytical tool in banks’ strategies. In recent years, researchers are developing new techniques for estimation. Financial performance evaluation of banks is a kind of multi-criteria decision making (MCDM) problem which has developed rapidly. It is very important for a firm to monitor a wide range of performance indicators in order to ensure that appropriate and timely decisions and plans can be made. Suitable performance measures can ensure that managers adopt a long-term perspective and allocate the company’s resources to the most effective activities. The aim of this study is to evaluate the financial performance model of Turkish Banks during 2012-2015 using forecasting (based on 2002-2011 data) methods and multi criteria decision techniques. As forecasting analysis tools, classical time series methods such as moving averages, exponential smoothing, Brown's single parameter linear exponential smoothing, Brown’s second-order exponential smoothing, Holt's two parameter linear exponential smoothing and decomposition methods applied to financial ratios data. After forecasting techniques Analytical Hierarchy Process (AHP) and Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) methodologies are used for the outranking of banks. This model is applied to a case study for the financial performance evaluation of 3 state banks (Ziraat Bank, Halk Bank and Vakıflar Bank); 9 private banks (Akbank; Anadolubank; Sekerbank; Tekstil Bank; Turkish Bank; Turk Ekonomi Bank; Garanti Bank; Is Bank and Yapı Kredi Bank) and 5 foreign banks (Denizbank; Eurobank Tekfen; Finans Bank; HSBC Bank and ING Bank) in Turkey. Financial performances of a bank is divided into ten groups including Capital Ratios, Balance Sheet Ratios, Assets Quality, Liquidity, Profitability, Income-Expenditure Structure, Share in Sector, Share in Group, Branch Ratios and Activity Ratios as described by the Banks Association of Turkey Tue, 01 Jan 2013 00:00:00 GMT https://www.um.edu.mt/library/oar/handle/123456789/70846 2013-01-01T00:00:00Z https://www.um.edu.mt/library/oar/handle/123456789/70845 Title: Globalization, regime-switching, and EU stock markets : the impact of the sovereign debt crises Authors: Ferreira, Nuno; Menezes, Rui; Bentes, Sónia Abstract: The most recent models learn over time, making the necessary adjustments to a new level of peaks or troughs, which enables the more accurate prediction of turning points. The Smooth Regression Model may be regarded as having a linear and a nonlinear component and may over time determine whether there is only a linear or nonlinear component or, in some cases, both. The present study focuses on the impact effect analysis of the European markets contamination by sovereign debt (particularly in Portugal, Spain, France and Ireland). The smooth transition regression approach applied in this study has proved to be a viable alternative for the analysis of the historical behavioural adjustment between interest rates and stock market indices. We found evidence in the crisis regime, i.e., large negative returns, especially in the case of Portugal, where we obtained the greatest nonlinear threshold adjustment between interest rates and stock market returns. Tue, 01 Jan 2013 00:00:00 GMT https://www.um.edu.mt/library/oar/handle/123456789/70845 2013-01-01T00:00:00Z https://www.um.edu.mt/library/oar/handle/123456789/70843 Title: The Seveso directives and their application to enterprise risk management Authors: Da Cruz, Manuel; Bentes, Sónia Abstract: Enterprise Risk Management is a relatively new concept which has emerged as a new paradigm for managing the portfolio of risks that face organizations. In this paper we explain what enterprise risk management is, how it differs from traditional risk management, what new skills are involved in this process and what advantages and opportunities this approach offers compared to prior techniques. Additionally, a relation with the Seveso Directives as a tool to manage risk is also provided. We conclude that Seveso Directives are an effective mechanism to achieve ERM objectives. Yet, this will only be accomplished by using Seveso activity as an improvement process rather than a compliance focused activity. Indeed, if applied with an eye beyond pure compliance, Seveso reports can deliver significant business opportunity. Tue, 01 Jan 2013 00:00:00 GMT https://www.um.edu.mt/library/oar/handle/123456789/70843 2013-01-01T00:00:00Z