Using of Statistical and Machine Learning Methods in Financial Markets

Please enable javascript in your browser.

Rostamzadeh, Mehrdad | 2020

340 Viewed

Type of Document: M.Sc. Thesis
Language: Farsi
Document No: 52591 (01)
University: Sharif University of Technology
Department: Industrial Engineering
Advisor(s): Kianfar, Farhad
Abstract:
The problem of stock price direction prediction is of great value among investors and researchers in the past decades. Even the smallest improvement in the performance of forecasting methods can lead to noticeable profit for investors. In this regard, in this research, a new method for filling the literature gap in the field of stock price direction forecasting is proposed. In the proposed method, two concepts of dynamics and model selection in dealing with data is investigated. Finally a predictive model is developed according to the two abovementioned concepts. Moreover, in this work, using a meta-learning approach one step towards making the prediction process automatic is taken. The developed method has been tested on the stocks of different companies. The obtained result indicates its outperformance against other benchmark models. In this study, the goal is to answer the problems of determining the input variables, ways of creating flexibility in predicting different data, methods of selecting the best predictive model among a pool of defined models and comparing the performance of the proposed dynamic model with the previously developed static models
Keywords:
Prediction ; Classifier ; Neural Network ; Financial Market ; Statistical Machine Learning ; Dynamic Ensemble Selection ; Metalearning

No TOC