Dataset Information

A deep learning framework for financial time series using stacked autoencoders and long-short term memory.

ABSTRACT: The application of deep learning approaches to finance has received a great deal of attention from both investors and researchers. This study presents a novel deep learning framework where wavelet transforms (WT), stacked autoencoders (SAEs) and long-short term memory (LSTM) are combined for stock price forecasting. The SAEs for hierarchically extracted deep features is introduced into stock price forecasting for the first time. The deep learning framework comprises three stages. First, the stock price time series is decomposed by WT to eliminate noise. Second, SAEs is applied to generate deep high-level features for predicting the stock price. Third, high-level denoising features are fed into LSTM to forecast the next day's closing price. Six market indices and their corresponding index futures are chosen to examine the performance of the proposed model. Results show that the proposed model outperforms other similar models in both predictive accuracy and profitability performance.

SUBMITTER: Bao W

PROVIDER: S-EPMC5510866 | biostudies-literature | 2017

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A deep learning framework for financial time series using stacked autoencoders and long-short term memory.

Bao Wei W Yue Jun J Rao Yulei Y

PloS one 20170714 7

The application of deep learning approaches to finance has received a great deal of attention from both investors and researchers. This study presents a novel deep learning framework where wavelet transforms (WT), stacked autoencoders (SAEs) and long-short term memory (LSTM) are combined for stock price forecasting. The SAEs for hierarchically extracted deep features is introduced into stock price forecasting for the first time. The deep learning framework comprises three stages. First, the stoc ...[more]

PMID: 28708865

Similar Datasets

Project description:BackgroundAn allergic reaction is the immune system's overreacting to a previously encountered, typically benign molecule, frequently a protein. Allergy reactions can result in rashes, itching, mucous membrane swelling, asthma, coughing, and other bizarre symptoms. To anticipate allergies, a wide range of principles and methods have been applied in bioinformatics. The sequence similarity approach's positive predictive value is very low and ineffective for methods based on FAO/WHO criteria, making it difficult to predict possible allergens.MethodThis work advocated the use of a deep learning model LSTM (Long Short-Term Memory) to overcome the limitations of traditional approaches and machine learning lower performance models in predicting the allergenicity of dietary proteins. A total of 2,427 allergens and 2,427 non-allergens, from a variety of sources, including the Central Science Laboratory and the NCBI are used. The data was divided 80:20 for training and testing purposes. These techniques have all been implemented in Python. To describe the protein sequences of allergens and non-allergens, five E-descriptors were used. E1 (hydrophilic character of peptides), E2 (length), E3(propensity to form helices), E4(abundance and dispersion), and E5 (propensity of beta strands) are used to make the variable-length protein sequence to uniform length using ACC transformation. A total of eight machine learning techniques have been taken into consideration.ResultsThe Gaussian Naive Bayes as accuracy of 64.14 %, Radius Neighbour's Classifier with 49.2 %, Bagging Classifier was 85.8 %, ADA Boost was 76.9 %, Linear Discriminant Analysis has 76.13 %, Quadratic Discriminant Analysis was 84.2 %, Extra Tree Classifier was 90%, and LSTM is 91.5 %.ConclusionAs the LSTM, has an AUC value of 91.5 % is regarded best in predicting allergens. A web server called ProAll-D has been created that successfully identifies novel allergens using the LSTM approach. Users can use the link https://doi.org/10.17632/tjmt97xpjf.1 to access the ProAll-D server and data.

Project description:Recent advances in deep learning for medical image segmentation demonstrate expert-level accuracy. However, application of these models in clinically realistic environments can result in poor generalization and decreased accuracy, mainly due to the domain shift across different hospitals, scanner vendors, imaging protocols, and patient populations etc. Common transfer learning and domain adaptation techniques are proposed to address this bottleneck. However, these solutions require data (and annotations) from the target domain to retrain the model, and is therefore restrictive in practice for widespread model deployment. Ideally, we wish to have a trained (locked) model that can work uniformly well across unseen domains without further training. In this paper, we propose a deep stacked transformation approach for domain generalization. Specifically, a series of n stacked transformations are applied to each image during network training. The underlying assumption is that the "expected" domain shift for a specific medical imaging modality could be simulated by applying extensive data augmentation on a single source domain, and consequently, a deep model trained on the augmented "big" data (BigAug) could generalize well on unseen domains. We exploit four surprisingly effective, but previously understudied, image-based characteristics for data augmentation to overcome the domain generalization problem. We train and evaluate the BigAug model (with n=9 transformations) on three different 3D segmentation tasks (prostate gland, left atrial, left ventricle) covering two medical imaging modalities (MRI and ultrasound) involving eight publicly available challenge datasets. The results show that when training on relatively small dataset (n = 10~32 volumes, depending on the size of the available datasets) from a single source domain: (i) BigAug models degrade an average of 11%(Dice score change) from source to unseen domain, substantially better than conventional augmentation (degrading 39%) and CycleGAN-based domain adaptation method (degrading 25%), (ii) BigAug is better than "shallower" stacked transforms (i.e. those with fewer transforms) on unseen domains and demonstrates modest improvement to conventional augmentation on the source domain, (iii) after training with BigAug on one source domain, performance on an unseen domain is similar to training a model from scratch on that domain when using the same number of training samples. When training on large datasets (n = 465 volumes) with BigAug, (iv) application to unseen domains reaches the performance of state-of-the-art fully supervised models that are trained and tested on their source domains. These findings establish a strong benchmark for the study of domain generalization in medical imaging, and can be generalized to the design of highly robust deep segmentation models for clinical deployment.

Dataset Information

A deep learning framework for financial time series using stacked autoencoders and long-short term memory.

Publications

A deep learning framework for financial time series using stacked autoencoders and long-short term memory.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets