International Research journal of Management Science and Technology

  ISSN 2250 - 1959 (online) ISSN 2348 - 9367 (Print) New DOI : 10.32804/IRJMST

Impact Factor* - 6.2311


**Need Help in Content editing, Data Analysis.

Research Gateway

Adv For Editing Content

   No of Download : 76    Submit Your Rating     Cite This   Download        Certificate

FORECASTING CPI-C OF INDIA USING ARIMA, SARIMA AND LSTM : A COMPARATIVE APPROACH

    2 Author(s):  RITU SINGH,SWAPNA SEN

Vol -  13, Issue- 3 ,         Page(s) : 174 - 186  (2022 ) DOI : https://doi.org/10.32804/IRJMST

Abstract

The time series data can be modelled using stochastic methods and Neural networks. This research aims to compare the three approaches ARIMA, SARIMA and LSTM in predicting the CPI-C of India. The method that this study will use in forecasting is Time Series Analysis. The approaches that are chosen are very different from each other. Every model has its own pros and cons.The objective of this research is to differentiate and to identify the model which provides the best predictions. This study will also aim at forecasting the future CPI-C of India using the best fit model. For evaluating the best fit, RMSE is taken as a parameter. The study is exploratory in nature. In the empirical analysis, we considered the time series consisting of CPI-C data obtained from the Reserve Bank of India's official website, between January 2013 - December 2021.

[1] Otu, O. A., Osuji, G. A., Opara, J., Mbachu, H. I., & Iheagwara, A. I. (2014). Application of Sarima models in modelling and forecasting Nigeria’s inflation rates. American Journal of Applied Mathematics and Statistics, 2(1), 16-28.
[2] Alonso, A. M., & Garcia-Martos, C. (2012). Time Series Analysis-Forecasting with ARIMA models. Universidad Carlos III de Madrid, Universidad Politecnica de Madrid.
[3] Urrutia, J. D. (2021). An Analysis on Forecasting Inflation Rate in the Philippines: A Recurrent Neural Network Method Approach. Turkish Journal of Computer and Mathematics Education (TURCOMAT), 12(3), 5297-5310.

*Contents are provided by Authors of articles. Please contact us if you having any query.






Bank Details