Show/Hide Menu
Hide/Show Apps
Logout
Türkçe
Türkçe
Search
Search
Login
Login
OpenMETU
OpenMETU
About
About
Open Science Policy
Open Science Policy
Open Access Guideline
Open Access Guideline
Postgraduate Thesis Guideline
Postgraduate Thesis Guideline
Communities & Collections
Communities & Collections
Help
Help
Frequently Asked Questions
Frequently Asked Questions
Guides
Guides
Thesis submission
Thesis submission
MS without thesis term project submission
MS without thesis term project submission
Publication submission with DOI
Publication submission with DOI
Publication submission
Publication submission
Supporting Information
Supporting Information
General Information
General Information
Copyright, Embargo and License
Copyright, Embargo and License
Contact us
Contact us
Performance of hybrid machine learning algorithms on financial time series data
Download
12626262.pdf
Date
2021-2-5
Author
Sayın, Merve Gözde
Metadata
Show full item record
This work is licensed under a
Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License
.
Item Usage Stats
1308
views
915
downloads
Cite This
Estimating stock indices that reflect the market has been an essential issue for a long time. Although various models have been studied in this direction, historically, statistical methods and then various machine learning methods have to introduced artificial intelligence into our lives. Related literature shows that neural networks and treebased models are mostly used. In this direction, in this thesis, four different models are examined. The first one is the most preferred neural network method for financial data called LSTM, and the second one is one of the most preferred tree-based models called XGBoost, and the third and the fourth models are the hybridizations of LSTM and XGBoost. Besides, these models have been applied to the total of nine stock market indexes, three from European markets, three from Asian and three from American markets, and the model that gives the best results is determined according to the Mean Absolute Scaled Error (MASE) evaluation criteria.
Subject Keywords
LSTM
,
XGBoost
,
Hybrid Models
,
Machine Learning
,
Stock Market Index
URI
https://hdl.handle.net/11511/89689
Collections
Graduate School of Applied Mathematics, Thesis
Suggestions
OpenMETU
Core
Application of r-vine copula method in Istanbul stock market data: A case study for the construction sector
Farnoudkia, Hajar; Purutçuoğlu Gazi, Vilda (Ankara Yıldırım Beyazıt Üniversitesi , 2020-12-01)
In the stock market, the relationship between the sectorial changes can be very informative in order to predict the changes in prices of assets from each sector. In order to understand these sectorial relations, various studies have been conducted. In one of the recent studies, the construction sector in Turkey was investigated in terms of its effect in other Turkish sectors since it is one of the leading sectors in Turkey and its assets have a significant impact in stock markets. Hereby, in this study we d...
Local volatility model applied to BIST30 european warrants: pricing and hedging
Kirazoğlu, Zekiye Sıla; Sezer, Ali Devin; Department of Financial Mathematics (2016)
One of the basic observations on pricing options is that the assumption of constant volatility does not agree with data and market price data gives a volatility smile that depends on maturities and strike prices. The first model that developed to be compatible with this observation is the local volatility model. The purpose of this work is to study the performance of the local volatility model on BIST30 warrants and compare it to the standard Black Scholes model. To estimate the local volatility model from d...
Do price limits help control stock price volatility?
Danışoğlu, Seza; Güner, Zehra Nuray (Springer Science and Business Media LLC, 2018-01-01)
On the negative side, price limits are criticized for increasing stock price volatility and hindering the price discovery process. On the positive side, price limits are argued to give panicky investors additional time to reassess their judgments and thus provide an opportunity for correcting the element of overreaction in pricing stocks. This study analyzes the effectiveness of price limits in Borsa Istanbul by utilizing a propensity-matched control sample in addition to the traditional benchmarks used in ...
How does the stock market volatility change after inception of futures trading?
Esen, İnci; Danışoğlu, Seza; Department of Financial Mathematics (2007)
As the trading volume in TURKDEX, the first and only options and futures exchange in Turkey, increases, it becomes more important to have an understanding of the effect of stock index futures trading on the underlying spot market volatility. In this respect, this thesis analyzes the effect of ISE-National 30 index futures contract trading on the underlying stocks’ volatility. In this thesis, spot portfolio volatility is decomposed into two components and this decomposition is applied to a single-factor retu...
An Analysis of momentum and mean reversion effects on equity indices
Özbilge, Armağan; Yolcu Okur, Yeliz; Nazlıben, Kamil Korhan; Department of Financial Mathematics (2015)
Momentum and mean-reversion effects have become very popular in finance literature for the last two decades since their presence can generate abnormal profit patterns by applying either relative strength or contrarian trading strategy accordingly. Even though there are some common factor explanations for return reversals, they might not provide the full picture for return persistence. In our theoretical framework, we analyse some of the well-known discrete time momentum studies including the initial one and...
Citation Formats
IEEE
ACM
APA
CHICAGO
MLA
BibTeX
M. G. Sayın, “Performance of hybrid machine learning algorithms on financial time series data,” M.S. - Master of Science, Middle East Technical University, 2021.