AUEB Library - Digital Repository

PYXIDA Institutional Repository
and Digital Library

Username
Password

Collections :	Ιδρυματικό Αποθετήριο ΟΠΑ / AUEB Institutional Repository Σχολή Οικονομικών Επιστημών / School of Economics Τμήμα Οικονομικής Επιστήμης / Department of Economics Μεταπτυχιακές Εργασίες / Postgraduate dissertations

Title :	Prediction of S&P 500 index movement using data mining techniques

Creator :	Michailidoy, Myrto-Christina

Contributor :	Kyriazidou, Ekaterini (Επιβλέπων καθηγητής) Tzavalis, Elias (Εξεταστής) Arvanitis, Stylianos (Εξεταστής) Athens University of Economics and Business, Department of Economics (Degree granting institution)

Type :	Text

Extent :	72p.

Language :	en

Abstract :	Predicting financial time series has proven extremely challenging due to their characteristics. There has been a vast number of researches investigating the predictability of financial variables from different aspects and by using different approaches. This study attempts to predict the direction of movement of S&P500 using models based on classification techniques; namely Logit, Linear Discriminant Analysis, Quadratic Discriminant Analysis, k-Nearest Neighbors, Support Vector Machines and Random Forest. The models developed are efficient, in the sense that any undetermined parameter is tuned using the Cross Validation technique. As inputs of the models, eleven technical indicators have been used and the data set is splitted into two sub-samples, the train and the test set. The performance of each model is assessed based on some evaluation measures, from which the best model is selected.

Abstract :

Predicting financial time series has proven extremely challenging due to their characteristics. There has been a vast number of researches investigating the predictability of financial variables from different aspects and by using different approaches. This study attempts to predict the direction of movement of S&P500 using models based on classification techniques; namely Logit, Linear Discriminant Analysis, Quadratic Discriminant Analysis, k-Nearest Neighbors, Support Vector Machines and Random Forest. The models developed are efficient, in the sense that any undetermined parameter is tuned using the Cross Validation technique. As inputs of the models, eleven technical indicators have been used and the data set is splitted into two sub-samples, the train and the test set. The performance of each model is assessed based on some evaluation measures, from which the best model is selected.

Subject :	Prediction S&P 500 Variance Inflation Factors (VIF) Financial variables Logit Linear Discriminant Analysis Quadratic Discriminant Analysis k-Nearest Neighbors Support Vector Machines (SVM) Random forest

Date :	28-02-2018

Licence :

File: Michailidou_2018.pdf

Type: application/pdf

Login