Stock price prediction using time series models

No Thumbnail Available
Authors
Chouksey, Shruti
Issue Date
2019
Degree
MSc in Data Analytics
Publisher
Dublin Business School
Rights
Items in eSource are protected by copyright. Previously published items are made available in accordance with the copyright policy of the publisher/copyright holder.
Abstract
This Thesis titled- “Stock price forecasting using time series models” focused on the comparison of the performance of time series models to predict the stock price for 5 banks. Forecasting and stock price analysis is important in finance and economics. Time series forecasting can be applied on any set of variables that change over time. For stocks or share prices, time series forecasting is common to track the price movement of the security over time. There is considerable past research work available on time series forecasting. In this thesis, a comparative study of time series forecasting using 3 models ARIMA (autoregressive integrated moving average), PROPHET and KERAS with LSTM (Long Short Term Memory) models has been explored. Historical stock price data was obtained from the National Stock Exchange (NSE) and used to build these models for comparative purposes. The results obtained reveal that all 3 models have strong potential for prediction and forecasting on the sourced historical data samples. All of the models performed better on larger data samples with LSTM best able to forecast seasonality.