An analysis of deep neural networks for predicting trends in time series data

Recently, a hybrid Deep Neural Network (DNN) algorithm, TreNet was proposed for predicting trends in time series data. While TreNet was shown to have superior performance for trend prediction to other DNN and traditional ML approaches, the validation method used did not take into account the sequential nature of time series data sets and did not deal with model update. In this research we replicated the TreNet experiments on the same data sets using a walk-forward validation method and tested our optimal model over multiple independent runs to evaluate model stability. We compared the performance of the hybrid TreNet algorithm, on four data sets to vanilla DNN algorithms that take in point data, and also to traditional ML algorithms. We found that in general TreNet still performs better than the vanilla DNN models, but not on all data sets as reported in the original TreNet study. This study highlights the importance of using an appropriate validation method and evaluating model stability for evaluating and developing machine learning models for trend prediction in time series data.

Automatic deep learning for trend prediction in time series data

Recently, Deep Neural Network (DNN) algorithms have been explored for predicting trends in time series data. In many real world applications, time series data are captured from dynamic systems. DNN models must provide stable performance when they are updated and retrained as new observations becomes available. In this work we explore the use of automatic machine learning techniques to automate the algorithm selection and hyperparameter optimisation process for trend prediction. We demonstrate how a recent AutoML tool, specifically the HpBandSter framework, can be effectively used to automate DNN model development. Our AutoML experiments found optimal configurations that produced models that compared well against the average performance and stability levels of configurations found during the manual experiments across four data sets.