Hybrid Forecasting: Combining LSTM and RNN with Classic ARIMA Models

Author(s)	Harpinder Kaur
Country	India
Abstract	The research showcases a comparative study of classical statistics-based and state-of-the-art deep learning time series forecasting models. Specifically, it investigates the performance of ARIMA (Auto Regressive Integrated Moving Average), “ARIMAX (ARIMA with exogenous variables), LSTM (Long Short-Term Memory) and a Hybrid LSTM-RNN (Recurrent Neural Network) approach. The purpose is to assess their performance to model and predict a complex, seasonal and non-linear sales trend with data from the stock market of Reliance Industries Limited (RIL). The datasets chosen for this study are well documented for their high seasonality and trend nature that is suitable for testing the forecast performance of the models. Exploratory data analysis and statistical diagnostics are preliminary steps in the analysis and include the Augmented Dickey-Fuller test for stationarity, autocorrelation and partial autocorrelation analysis, as well as Time Series decomposition into trend, seasonal and residual components. These first steps assist in the model's methodology and give some indication of the long-term modelling trends in the data.
Keywords	ARIMA, ARIMAX, LSTM, Hybrid LSTM-RNN, RMSE, MAE
Field	Mathematics
Published In	Volume 8, Issue 3, May-June 2026
Published On	2026-06-09

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research