FORECASTING BIST 100 INDEX USING ARTIFICIAL NEURAL NETWORKS (ANN) AND ADAPTIVE NEURO FUZZY INFERENCE SYSTEM (ANFIS) METHOD
Journal of Applied Research in Finance and Economics
View Archive InfoField | Value | |
Title |
FORECASTING BIST 100 INDEX USING ARTIFICIAL NEURAL NETWORKS (ANN) AND ADAPTIVE NEURO FUZZY INFERENCE SYSTEM (ANFIS) METHOD
|
|
Creator |
YILMAZ, Tayfun
KILIÇ, Bayram |
|
Description |
The purpose of this study is to compare forecasting performance of Artificial Neural Networks (ANN) and Adaptive Neuro Fuzzy Inference System (ANFIS) methods. In this study, BIST 100 index was forecasted with predetermined independent variables using ANN and ANFIS methods and then the forecasting performances of these two methods were compared. The data set of this study includes 2040 weekly-frequency data covering the period from February 11, 2011 to December 25, 2015. The empirical findings indicated that the estimation performances of these two models are quite close to each other. In addition, ANN technique was found to present a superior forecasting performance when compared to ANFIS technique.
|
|
Publisher |
Journal of Applied Research in Finance and Economics
|
|
Date |
2016-11-30
|
|
Type |
info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion Peer-reviewed Article |
|
Format |
application/pdf
|
|
Identifier |
http://www.jarfe.org/index.php/jarfe/article/view/10
|
|
Source |
Journal of Applied Research in Finance and Economics; Vol 2 No 1 (2016): Issue 2-1; 18-27
2458-8083 |
|
Language |
eng
|
|
Relation |
http://www.jarfe.org/index.php/jarfe/article/view/10/10
|
|
Rights |
Copyright (c) 2016 Journal of Applied Research in Finance and Economics
|
|