Forecasting Stock Market Indices Using Padding-Based Fourier Transform Denoising and Time Series Deep Learning Models.
Donghwan SongAdrian Matias Chung BaekNamhun KimPublished in: IEEE Access (2021)
Keyphrases
- learning models
- fourier transform
- stock market
- denoising
- financial time series
- frequency domain
- short term
- garch model
- image denoising
- investment strategies
- stock price
- long term
- machine learning
- stock exchange
- loss function
- image processing
- learning algorithm
- spatial domain
- machine learning algorithms
- learning tasks
- financial data
- conditional random fields
- stock data
- wavelet domain
- signal processing
- learning problems
- natural images
- semi supervised learning
- exchange rate
- classification models
- financial markets
- graphical models
- multivariate time series
- higher order
- semi supervised
- portfolio optimization
- unsupervised learning
- prediction model
- feature set
- supervised learning
- pairwise
- training data
- non stationary