Classification Of BIST -100 Index’ Changes Via Machine Learning Methods

Filiz, Enes; Öz, Ersoy

doi:10.14780/muiibd.329913

Cited by 7 publications

(4 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…According to the results, average DPA value is about 50% and DPA values change between 38% (KCHOL) and 70% (ARCLK). These results are on par with the results of the studies for stock movement prediction of BIST30 companies in the literature (Filiz & Öz, 2017; Özçalıcı, 2016). The results also show that our prediction model predicts monthly returns with an average RMS‐R value of 0.12; while the best RMS‐R value is 0.07 (ARCLK) and the worst RMS‐R value is 0.32 (BIMAS).…”

Section: Resultssupporting

confidence: 83%

A portfolio construction framework using LSTM‐based stock markets forecasting

Yıldız

Yildiz

2020

Int. J Fin Econ

View full text Add to dashboard Cite

A novel framework that injects future return predictions into portfolio constructionstrategies is proposed in this study. First, a long-short-term-memory (LSTM) model is trained to learn the monthly closing prices of the stocks.Then these predictions are used in the calculation of portfolio weights. Five different portfolio construction strategies are introduced including modifications to smart-beta strategies. The suggested methods are compared to a number of baseline methods, using the stocks of BIST30 Turkey index. Our strategies yield a very high mean annualized return (25%) which is almost 50% higher than the baseline approaches. The mean Sharpe ratio of our strategies is 0.57, whereas the compared methods' are 0.29 and −0.32. Comprehensive analysis of the results demonstrates that utilizing predicted returns in portfolio construction enables a significant improvement on the performance of the portfolios.

show abstract

Section: Resultssupporting

confidence: 83%

A portfolio construction framework using LSTM‐based stock markets forecasting

Yıldız

Yildiz

2020

Int. J Fin Econ

View full text Add to dashboard Cite

show abstract

“…The entire data set is thus broken into pieces k times and used as testing sets. Then the mean of all the testing is computed to result in classification measures (Filiz & Öz, 2017).…”

Section: Data Setmentioning

confidence: 99%

Finding the Best Algorithms and Effective Factors in Classification of Turkish Science Student Success

Filiz

Öz

2019

JBSE

Self Cite

View full text Add to dashboard Cite

Educational Data Mining (EDM) is an important tool in the field of classification of educational data that helps researchers and education planners analyse and model available educational data for specific needs such as developing educational strategies. Trends International Mathematics and Science Study (TIMSS) which is a notable study in educational area was used in this research. EDM methodology was applied to the results of TIMSS 2015 that presents data culled from eighth grade students from Turkey. The main purposes are to find the algorithms that are most appropriate for classifying the successes of students, especially in science subjects, and ascertaining the factors that lead to this success. It was found that logistic regression and support vector machines – poly kernel are the most suitable algorithms. A diverse set of features obtained by feature selection methods are “Computer Tablet Shared”, “Extra Lessons Last 12 Month”, “Extra Lessons How Many Month”, “How Far in Education Do You Expect to Go”, “Home Educational Resources”, and “Student Confident in Science” and these features are the most effective features in science success. Keywords: classification algorithms, educational data mining, eighth grade, science success, TIMSS 2015.

show abstract

“…The success of the estimation results obtained was evaluated with the R 2 metric and values between 0.95-0.98 were reached. In the [7], the direction of change of the BIST 50 index was estimated with artificial neural networks (ANN), KNN, Naive Bayes and C4.5 decision tree models. The success of the estimation results obtained was evaluated with the classification accuracy, and the highest value was obtained as 92.71% with the C4.5 decision tree model.…”

Section: The Related Studiesmentioning

confidence: 99%

Sector-Based Stock Price Prediction with Machine Learning Models

KOCAOĞLU

TURGUT

Konyar

2022

Sakarya University Journal of Computer and Information Sciences

View full text Add to dashboard Cite

Stock price prediction is an important topic for investors and companies. The increasing effect of machine learning methods in every field also applies to stock forecasting. In this study, it is aimed to predict the future prices of the stocks of companies in different sectors traded on the Borsa Istanbul (BIST) 30 Index. For the study, the data of two companies selected as examples from each of the holding, white goods, petrochemical, iron and steel, transportation and communication sectors were analyzed. In the study, in addition to the share analysis of the sectors, the price prediction performances of the machine learning algorithm on a sectoral basis were examined. For these tests, XGBoost, Support Vector Machines (SVM), K-nearest neighbors (KNN) and Random Forest (RF) algorithms were used. The obtained results were analyzed with mean absolute error (MAE), mean absolute percent error (MAPE), mean squared error (MSE), and R2 correlation metrics. The best estimations on a sectoral basis were made for companies in the Iron and Steel and Petroleum field. One of the most important innovations in the study is the examination of the effect of current macro changes on the forecasting model. As an example, the effect of the changes in the Central Bank Governors, which took place three times in the 5-year period, on the forecast was investigated. The results showed that the unpredictable effects on the policies after the change of Governors also negatively affected the forecast performance

show abstract

Classification Of BIST -100 Index’ Changes Via Machine Learning Methods

Abstract: The changes in

Cited by 7 publications

References 5 publications

A portfolio construction framework using LSTM‐based stock markets forecasting

A portfolio construction framework using LSTM‐based stock markets forecasting

Finding the Best Algorithms and Effective Factors in Classification of Turkish Science Student Success

Sector-Based Stock Price Prediction with Machine Learning Models

Contact Info

Product

Resources

About