caa a22 4500 605477892 CHVBK 20210128100403.0 cr unu---uuuuu 210128e20151001xx s 000 0 eng 10.1007/s10994-014-5447-y doi (NATIONALLICENCE)springer-10.1007/s10994-014-5447-y Measuring the accuracy of currency crisis prediction with combined classifiers in designing early warning system [Elektronische Daten] [Nor Ramli, Mohd Ismail, Hooy Wooi] Is the prediction accuracy affected by the method used in the ensemble of the classifiers? This paper is a sequel of our experiment in order to find an answer for such question. Previously, we had conducted an experiment by using single classifiers in the machine learning against traditional statistical methods. The results showed that single classifiers in machine learning perform well compared to the traditional statistical methods. Still, we believe that there is another way to increase the prediction accuracy of these classifiers. In this paper, we conducted another experiment by combining these classifiers in predicting currency crisis of 25 countries. The combined classifiers are support vector machine with k-nearest neighbor, logistic regression with k-nearest neighbor and finally LADTree with k-nearest neighbor. These three combined classifiers are tested on 13 chosen macroeconomic indicators which the data is taken from first quarter 1980 to third quarter 2012. The results of this experiment showed that these three different combined classifiers averagely have same higher accuracy and quite comparable. Our proposed method, nearest neighbor tree has the highest area under ROC curve number among these three combined classifiers although in terms of computational time it took longer running times than the others. The Author(s), 2014 Machine learning nationallicence Combined classifiers nationallicence Currency crisis nationallicence Early warning system nationallicence k-Nearest neighbor method nationallicence Ramli Nor School of Mathematical Sciences, Universiti Sains Malaysia, Penang, Malaysia aut Ismail Mohd School of Mathematical Sciences, Universiti Sains Malaysia, Penang, Malaysia aut Wooi Hooy School of Management, Universiti Sains Malaysia, Penang, Malaysia aut Machine Learning Springer US; http://www.springer-ny.com 101/1-3(2015-10-01), 85-103 0885-6125 101:1-3<85 2015 101 10994 https://doi.org/10.1007/s10994-014-5447-y text/html Onlinezugriff via DOI BK010053 XK010053 XK010000 Metadata rights reserved Springer special CC-BY-NC licence nationallicence 1 research-article jats NATIONALLICENCE NATIONALLICENCE NL-springer NATIONALLICENCE 856 40 https://doi.org/10.1007/s10994-014-5447-y text/html Onlinezugriff via DOI NATIONALLICENCE 700 1- Ramli Nor School of Mathematical Sciences, Universiti Sains Malaysia, Penang, Malaysia aut NATIONALLICENCE 700 1- Ismail Mohd School of Mathematical Sciences, Universiti Sains Malaysia, Penang, Malaysia aut NATIONALLICENCE 700 1- Wooi Hooy School of Management, Universiti Sains Malaysia, Penang, Malaysia aut NATIONALLICENCE 773 0- Machine Learning Springer US; http://www.springer-ny.com 101/1-3(2015-10-01), 85-103 0885-6125 101:1-3<85 2015 101 10994