Multiscale Model Selection for High-Frequency Financial Data of a Large Tick Stock by Means of the Jensen-Shannon Metric.
Gianbiagio CuratoFabrizio LilloPublished in: Entropy (2014)
Keyphrases
- model selection
- high frequency
- financial data
- stock market
- multiscale
- stock price
- wavelet transform
- low frequency
- financial information
- cross validation
- wavelet coefficients
- hyperparameters
- subband
- high resolution
- parameter estimation
- machine learning
- mixture model
- image processing
- variable selection
- selection criterion
- short term
- metric learning
- edge detection
- feature selection
- maximum likelihood
- image fusion
- training set
- computational complexity
- feature extraction