Borsa analizi, geleceğe yönelik tahminler yapmak için finansal, politik ve sosyal göstergeleri göz önünde bulundurarak borsayı inceler ve değerlendirir. Büyük veri ve derin öğrenme teknolojilerindeki gelişmelerin çığır açan sonuçları, araştırmacıların ve endüstrinin dikkatini bilgisayar destekli borsa analizine çekmektedir. Geleneksel makine öğrenimi ve derin öğrenme modellerini kullanarak borsa analizi konusunda çeşitli çalışmalar bulunmaktadır. Bu çalışmada, temel model olarak Otoregresif Entegre Hareketli Ortalama (ARIMA) yöntemini tekrarlayan sinir ağlarının üç farklı modeliyle karşılaştırılmıştır; Uzun Kısa Süreli Bellek (Long Short Term Memory- LSTM) ağları, Geçitli Tekrarlayan Birim (Gated Recurrent Unit- GRU), dikkat katmanlı LSTM modeli. Bu çalışmada literatürdeki diğer çalışmalardan farklı olarak 28 tane finansal indikatör kullanılarak Borsa İstanbul verileri üzerinde gün içi tahminler yaparken dört farklı modelin sonuçları karşılaştırılmıştır. İstatistiksel ve doğrusal bir model olan ARIMA, zaman serileri tahmini için doğrusal olmayan RNN modelleri ile karşılaştırılmıştır ancak 3 sinir ağı modelinden de yüksek ortalama hata oranına sahip olduğu görülmüştür. LSTM sonuçları GRU modeline çok yakın olsa da GRU diğerlerinden biraz daha iyi performans göstermektedir. Dikkat mekanizmalı sinir ağı diğer temel sinir ağlarından daha iyi sonuç vermemektedir.
Otoregresif Entegre Hareketli Ortalama BIST Geçitli Tekrarlayan Birim Modeli Uzun Kısa Süreli Bellek Modeli
The stock market analysis examines and evaluates the stock market by considering the financial, political, and social indicators to make future predictions. Breakthrough results of advancements in big data and deep learning technologies attract the attention of researchers and
traders to computer-assisted stock market analysis. There are several studies on stock market analysis using conventional machine learning and deep learning models. In this paper, we used Autoregressive Integrated Moving Average (ARIMA) as a base model and compared it with three different models of recurrent neural networks: Long Short-Term Memory (LSTM) networks, Gated Recurrent Unit (GRU), LSTM with an attention layer model. While making intraday forecasts on Borsa Istanbul data, the results of four different models have been compared. The statistical model, ARIMA, is used as a baseline model for comparison with neural networks, but it has higher mean absolute error than other neural network models. Even though the LSTM results are very close to the GRU model, GRU slightly outperforms the others. The attention neural network model does not give better results than other basic neural networks.
Primary Language | Turkish |
---|---|
Subjects | Engineering |
Journal Section | Makaleler(Araştırma) |
Authors | |
Early Pub Date | October 22, 2023 |
Publication Date | November 20, 2023 |
Published in Issue | Year 2023 Volume: 16 Issue: 2 |
Article Acceptance
Use user registration/login to upload articles online.
The acceptance process of the articles sent to the journal consists of the following stages:
1. Each submitted article is sent to at least two referees at the first stage.
2. Referee appointments are made by the journal editors. There are approximately 200 referees in the referee pool of the journal and these referees are classified according to their areas of interest. Each referee is sent an article on the subject he is interested in. The selection of the arbitrator is done in a way that does not cause any conflict of interest.
3. In the articles sent to the referees, the names of the authors are closed.
4. Referees are explained how to evaluate an article and are asked to fill in the evaluation form shown below.
5. The articles in which two referees give positive opinion are subjected to similarity review by the editors. The similarity in the articles is expected to be less than 25%.
6. A paper that has passed all stages is reviewed by the editor in terms of language and presentation, and necessary corrections and improvements are made. If necessary, the authors are notified of the situation.
. This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.