Forecasting and Analyzing Predictors of Inflation Rate: Using Machine Learning Approach

Pijush Kanti Das () and Prabir Kumar Das
Additional contact information
Pijush Kanti Das: Indian Institute of Foreign Trade
Prabir Kumar Das: Indian Institute of Foreign Trade

Journal of Quantitative Economics, 2024, vol. 22, issue 2, No 10, 493-517

Abstract: Abstract In this study, we investigate and apply the models from the machine learning (ML) paradigm to forecast the inflation rate. The models identified are ridge, lasso, elastic net, random forest, and artificial neural network. We carry out the analysis using a data set with 56 features of 132 monthly observations from January 2012 to December 2022. The random forest (RF) model can forecast the inflation rate with greater accuracy than other ML models. A comparison to benchmark econometric models like auto-regressive integrated moving average demonstrates the superior performance of the RF model. Moreover, nonlinear ML models are proven to be more successful than a linear ML or time series models and this is mostly due to the unpredictability and interactions of variables. It indicates that the significance of nonlinear structures for forecasting inflation is important. Furthermore, the ML models outweigh the benchmark econometric model in forecasting the undulations due to the COVID-19 impact. The findings in this study support the benefit of applying ML models to forecast the inflation rate. Even without considering the sporadicity of pandemic, nonlinear model like artificial neural network (ANN) outweighs other models. Additionally, the ML models like RF and ANN model yield variable importance measures for each explanatory variable. ML models shows capability to not only better forecasting but also able to provide the insight regarding the covariates for improved forecasting results and policy prescriptions.

Keywords: Inflation; Forecasting; Random forest; Artificial neural networks; Machine learning (search for similar items in EconPapers)
JEL-codes: C4 C5 C8 (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://link.springer.com/10.1007/s40953-024-00384-z Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:jqecon:v:22:y:2024:i:2:d:10.1007_s40953-024-00384-z

Ordering information: This journal article can be ordered from
http://www.springer.com/economics/journal/40953

DOI: 10.1007/s40953-024-00384-z

Access Statistics for this article

Journal of Quantitative Economics is currently edited by Dilip Nachane and P.G. Babu

More articles in Journal of Quantitative Economics from Springer, The Indian Econometric Society (TIES) Contact information at EDIRC.
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().