Assessment Of Artificial Neural Network To Improve Hidden Markov Model For Financial Data

Download

10486076.pdf

Date

2022-7-27

Author

Aydoğan Kılıç, Dilek

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

315
views

177
downloads

The aim of this thesis is to eliminate the possible weaknesses of HMMs, which is a successful statistical model that is frequently used in time series modeling. Depending on the selection of the initial parameters of the HMMs, RNN is used as a solution to the failure to reach the global maximum, and it is aimed to benefit from the classification power of this method. The hybrid model, which is developed with this motivation, is built in a way that is suitable for use in non-categorical data, contrary to the version generally used in the literature. In this thesis, the hybrid model, which is effective in the development of speech recognition in the literature, is reconstructed and applied to financial data. Additionally, a multivariate comparison is conducted in order to identify the effect of the other variables in the model. Therefore, apart from univariate models, bivariate and trivariate models are also constructed. Moreover, classical HMM and RNN are applied and compared with the Hybrid model results. The applications use daily closing prices for the S&P 500 and Nasdaq and daily EUR/USD exchange rates from 2000 to 2021. In comparison to the single HMM and RNN methods, the accuracy in forecasting is significantly increased.

Subject Keywords

Hidden Markov Model, Neural Network, stock prices, forecasting, EUR/USD exchange rates

URI

https://hdl.handle.net/11511/98552

Collections

Graduate School of Applied Mathematics, Thesis

Suggestions

OpenMETU
Core

Overconfidence and bubbles in experimental asset markets Şahin, Serkan; Küçükkaya, Halit Engin; Yılmaz, Özlem; Department of Business Administration (2013) The aim of this study is to investigate uncertainty levels of industries and explore those financial ratios that have the highest information content in determining the set of industry characteristics and use the most informative ratios selected in developing industry specific financial distress models. First, we employ factor analysis to determine the set of ratios that are most informative in specified industries. Second, we use entropy method as a Multiple Attribute Decision Making Model, to measure the ...
Hybrid wavelet-neural network models for time series data Kılıç, Deniz Kenan; Uğur, Ömür; Department of Financial Mathematics (2021-3-3) The thesis aims to combine wavelet theory with nonlinear models, particularly neural networks, to find an appropriate time series model structure. Data like financial time series are nonstationary, noisy, and chaotic. Therefore using wavelet analysis helps better modeling in the sense of both frequency and time. S&P500 (∧GSPC) and NASDAQ (∧ IXIC) data are divided into several components by using multiresolution analysis (MRA). Subsequently, each part is modeled by using a suitable neural network structure. ...
Pricing spread and basket options under markov-modulated models Kozpınar, Sinem; Uğur, Ömür; Eksi Altay, Zehra; Department of Financial Mathematics (2018) This thesis first aims to study the evaluation of spread and basket options under the classical Markov-modulated framework, for which a transition in the Markov process leads to a switch in the model parameters. In this regard, we provide approximations to the exact option prices based on ideas from the literature without regime switching. We start with pricing spread options when risky assets follow Markov-modulated geometric Brownian motions (MMGBMs). In this context, we focus on the regime-switching vers...
Identification of coupled systems of stochastic differential equations in finance including investor sentiment by multivariate adaptive regression splines Kalaycı, Betül; Weber, Gerhard Wilhelm; Department of Financial Mathematics (2017) Stochastic Differential Equations (SDEs) rapidly become the most well-known format in which to express such diverse mathematical models under uncertainty such as financial models, neural systems, micro-economic systems, and human behaviour. They are one of the main methods to describe randomness of a dynamical model today. In a financial system, different kinds of SDEs have been elaborated to model various financial assets. On the other hand, economists have conducted research on several empirical phenomena...
Critical realism as a rival methodology for institutional economics Gürpınar, Erkan; Özveren, Eyüp; Department of Economics (2008) The aim of the thesis is to clarify the methodology of the original institutional economics and then to evaluate the current attempts to utilize critical realism as a superior methodology for it. After sketching the historical background of the discussions surrounding methodology of science and 19th century economic methodology, the thesis separately analyses the methodology of institutional economics and critical realist stance in the philosophy of science. A critical discussion of the subject matter reach...

Citation Formats

D. Aydoğan Kılıç, “Assessment Of Artificial Neural Network To Improve Hidden Markov Model For Financial Data,” Ph.D. - Doctoral Program, Middle East Technical University, 2022.