Jelena Stanković

Investment Strategy Optimization Using Technical Analysis and Predictive Modeling in Emerging Markets

Abstract This research examines the efficacy of technical analysis and predictive modeling in defining the optimal strategy for investing in the stocks indices of emerging markets. Trading strategies are set regarding different technical indicators based on moving averages and volatility of the value and returns on stock indices. Simple trading rules are generated using two moving averages – a long period and a short period moving average, and Moving Average Convergence-Divergence (MACD) and Relative Strength Index (RSI). Selected technical indicators are used as features in defining predictive model based on Least Squares Support Vector Machines (LS-SVMs). A LS-SVM classifier has been used in order to predict trend of the stock indices’ value whereby the obtained outputs of the LS-SVM model are binary signals that can be used for defining the trading strategy. Comparing the results obtained from traditional statistical methods for predicting the trend of financial series and proposed LS-SVM model, it can be concluded that machine learning techniques capture the non-linear models which are dominant in the financial markets in more adequate way. Outperforming the results of Buy & Hold strategy and technical trading strategies, application of LS-SVM in decision making process on investing on the financial market significantly can contribute to maximization of profitability on investment.

Stock Market Trend Prediction Based on the LS-SVM Model Update Algorithm

The paper proposes a trend prediction model based on an incremental training set update scheme for the BELEX15 stock market index using the Least Squares Support Vector Machines (LS-SVMs) for classification. The basic idea of this updating approach is to add the most recent data to the training set, as become available. In this way, information from new data is taken into account in model training. The test results indicate that the suggested model is suitable for short-term market trend prediction and that prediction accuracy significantly increases after the training set has been updated with new information.

Multi-criteria approach in evaluating contribution of social entrepreneurship to the employment of socially-excluded groups

The paper explores the potential impact of social enterprises on social exclusion. In particular, the role of social enterprises in labour market integration of socially excluded individuals is analysed within the existing theoretical and policy discourses of exclusion. Taking into account the difficulties in measuring the social impact of social enterprises, our study contributes to the quantitative literature on the performance of social enterprises, taking the number of integrated people as a measure of the impact on social exclusion. The research is based on data on the employment of marginalized groups, derived from a unique dataset collected by a recently conducted survey of social enterprises in Serbia. The original methodological framework combines statistical methods and multi-criteria decision making model, in order to evaluate the contribution of different types of enterprises to the employment of excluded individuals. Weights determination for the MCDM model is performed using entropy while TOPSIS method is applied for the ranking of the types of social enterprises according to the employment of socially excluded categories. The results indicate that enterprises for employment of persons with disabilities, citizens’ associations and cooperatives in Serbia contribute the most in integrating the socially excluded.

Stock market trend prediction using AHP and weighted kernel LS-SVM

Nowadays, stock market trend prediction represents a challenging subject both in terms of the choice of the prediction model and in terms of constructing the set of features that model will use for prediction. To address both of these aspects, we propose a feature ranking and feature selection approach in combination with weighted kernel least squares support vector machines (LS-SVMs). We introduce the analytic hierarchy process (AHP) into the stock market and propose evaluation criteria which provide the prediction model with relevant knowledge of the underlying processes of the studied stock market. The feature weights obtained by the AHP method are used for feature ranking and selection, and used with the LS-SVMs through a weighted kernel. The test results indicate that the proposed model outperforms the benchmark models. In addition, the set of feature weights obtained by the proposed approach can also independently be incorporated into other kernel-based learners.

Interaction Between Competitiveness and Innovation: Evidence from South-Eastern European Countries

The traditional doctrine of classical economic theory mostly considers economic growth and competitiveness at the level of the national economy. The problem of competitiveness within this stream of economic theory is mainly approached as the need to find the answers why some countries develop faster and become richer than others, i.e. how an economy gains the ability of sustainable growth, which makes it more competitive than others. In contrast to this point of view, the neoclassical economic theory shifts the focus of the research of growth and competitiveness from the national economy to the field of resource allocation and enterprise efficiency. The logic is simple—competitive enterprises make competitive economy and vice versa, competitive economy creates an environment that encourages the competitiveness of enterprises. The research concept of this paper is based on the synthesis of these two approaches.

Determination of Expert Group Preferences in the Multi-Criteria Model for the Analysis of Local Economic Environment

Abstract Local economic environment is characterised by a range of economic, social, political and demographic parameters, based on which we can perform its analysis. Heterogeneity of relevant characteristics of the local economic environment imposes multiple criteria analysis as one of the suitable tools for the evaluation. Assessment of local economic environment often falls within the scope of group decision-making, as it is usually performed on the basis of an analysis of preferences of economic subjects or relevant experts on the issue of the economic environment at the local level. Regardless of whether it is based on economic subjects or expert group, in order to form a multi-criteria model, it is necessary to generate preferences of individuals into a single weight coefficient, which shows groups’ preference on the importance of each criterion. The subject of this paper is determination of weight coefficients in the multi-criteria model for the analysis of local economic development based on the preferences from a group of experts, by applying adequate statistical tools, and then by ranking local governments according to the quality of business environment perceived by the expert group.In addition to descriptive statistics and testing the significance of differences, in the paper is applied multi-criteria method Simple Additive Weights - SAW.

Enhancing local economic development using collective intelligence

This paper focus on collecting opinions and suggestions by entrepreneurs in order to enhance local economic development. Questions are expressed in the form of crowdsourcing pool, and answers are clustered by similarity. The frequency of similar answers obtained by the same or similar questions may point to obstacles for economic development that local government institutions should solve.

Decision-making under uncertainty – the integrated approach of the AHP and Bayesian analysis

In situations where it is necessary to perform a large number of experiments in order to collect adequate statistical data which require expert analysis and assessment, there is a need to define a model that will include and coordinate statistical data and experts’ opinions. This article points out the new integrated application of the Analytic Hierarchy Process (AHP) and Bayesian analysis, in the sense that the Bayes’ formula can improve the accuracy of input data for the Analytical Hierarchy Process, and vice versa, AHP can provide objectified inputs for the Bayesian formula in situations where the statistical estimates of probability are not possible. In this sense, the AHP can be considered as the Bayesian process that allows decision-makers to objectify their decisions and formalise the decision process through pairwise comparison of elements.

Data preparation for modeling predictive analizes in the field of basic property insurance risks

This paper describes the process of preparing available data in the field of basic property insurance risks, in order to use it in the further researches for modeling the probability distribution of the occurrence of insured events that cause extremely large damages. The paper also describes a descriptive and exploratory data analysis on data set for the period between 2001 and 2011 on the territory of the Republic of Serbia. The purpose of data preprocessing is to reduce the amount of data to be analyzed, so as to improve the quality of the analysis. Process steps include data cleaning and that data transformation, in order to increase speed and prevent the domination of certain attributes in the application of algorithms for classification and clustering.