Downloads: 2 | Views: 42 | Weekly Hits: ⮙2 | Monthly Hits: ⮙2
Research Paper | Financial Engineering | India | Volume 13 Issue 4, April 2024 | Rating: 5 / 10
Deep Learning for Financial Time Series using Long Short-Term Memory Model
Abstract: In this paper, we will be using the LSTM model to predict the next day's updtrend for financial time series. The idea here is to analyze whether LSTM is able to predict the sign of daily returns based on adjusted close prices. Thus, it is a classification problem. I begin with the problem statement of this paper i.e. the use of LSTM model for time series prediction. The data has been cleaned to remove any null values and use the linear interpolation method in case there are any such values. The pandas TA library has been used in order to generate features. Importing all the stategies present in the library, I have used the BorutaPy method of feature selection as it is a robust method that provides us with the relevant feature ranking that can be used for our model.
Keywords: Deep Learning, Time Series, Memory Model
Edition: Volume 13 Issue 4, April 2024,
Pages: 1944 - 1972