A "time sequence analysis" is a particular method for looking at a group of data points gathered over a long period of time. Instead of merely randomly or infrequently, time series analyzers gather information from data points over a predetermined length of time at scheduled times. But this kind of research requires more than just accumulating data over time. Data in time series may be analyzed to illustrate how variables change over time, which makes them different from other types of data. To put it another way, time is a crucial element since it demonstrates how the data changes over the period of the information and the outcomes. It offers a predetermined architecture of data dependencies as well as an extra data source. Time Series forecasting is a crucial field in deep learning because many forecasting issues have a temporal component. A time series is a collection of observations that are made sequentially across time. In this study, we examine distinct machine learning, deep learning and ensemble model algorithms to predict Nike stock price. We are going to use the Nike stock price data from January 2006 to January 2018 and make predictions accordingly. The outcome demonstrates that the hybrid LSTM-GRU model outperformed the other models in terms of performance.
Tripathy, Nrusingha; Parida, Surabi; and Nayak, Subrat Kumar
"Forecasting Stock Market Indices Using Gated Recurrent Unit (GRU) Based Ensemble Models: LSTM-GRU,"
International Journal of Computer and Communication Technology: Vol. 9:
1, Article 11.
Available at: https://www.interscience.in/ijcct/vol9/iss1/11