Forecasting of Inflation Based on Univariate and Multivariate Time Series Models: An Empirical Application

Hasnain Iftikhar (), Faridoon Khan, Paulo Canas Rodrigues, Abdulmajeed Atiah Alharbi and Jeza Allohibi
Additional contact information
Hasnain Iftikhar: Department of Statistics, Quaid-i-Azam University, Islamabad 45320, Pakistan
Faridoon Khan: Department of Creative Technology, Faculty of Computing and AI, Air University, Islamabad 44000, Pakistan
Paulo Canas Rodrigues: Department of Statistics, Federal University of Bahia, Salvador 40170-110, Brazil
Abdulmajeed Atiah Alharbi: Department of Mathematics, Faculty of Science, Taibah University, Al-Madinah Al-Munawara 42353, Saudi Arabia
Jeza Allohibi: Department of Mathematics, Faculty of Science, Taibah University, Al-Madinah Al-Munawara 42353, Saudi Arabia

Mathematics, 2025, vol. 13, issue 7, 1-18

Abstract: Maintaining stable prices is one of the goals of monetary policy makers. Since its formation, inflation has been a key issue and priority for every Pakistani government; it is a fundamental macroeconomic variable that plays a significant role in a nation’s economic progress and development. This research investigates the predictive capabilities of different univariate and multivariate models. The study considers autoregressive models, autoregressive neural networks, autoregressive moving average models, and other nonparametric autoregressive models within the univariate category. In contrast, the multivariate models include factor models that utilize Minimax Concave Penalty, Elastic-Smoothly Clipped Absolute Deviation, Principal Component Analysis, and Partial Least Squares. We conducted an empirical analysis using a well-established macroeconomic dataset from Pakistan. This dataset covers the period from January 2013 to December 2020 and consists of 79 variables recorded at that frequency. To evaluate the forecasting accuracy of the models for multiple steps ahead in the post-sample period, an analysis was performed using data extracted from January 2013 to February 2019 for model estimation and then another set from March 2019 to December 2020. The predictability of the univariate models following the sample period is compared with that of the multivariate models using statistical accuracy measurements, specifically root mean square error and mean absolute error. Additionally, the Diebold–Mariano test has been employed to evaluate the accuracy of the average errors statistically. The results indicated that the factor approach based on Partial Least Squares delivers significantly more effective outcomes than its competing methods.

Keywords: fat big data; inflation forecasting; univariate and multivariate time series forecasting models; multi-step ahead forecasting; financial time series analysis; decision making (search for similar items in EconPapers)
JEL-codes: C (search for similar items in EconPapers)
Date: 2025
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2227-7390/13/7/1121/pdf (application/pdf)
https://www.mdpi.com/2227-7390/13/7/1121/ (text/html)

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:gam:jmathe:v:13:y:2025:i:7:p:1121-:d:1623158

Access Statistics for this article

Mathematics is currently edited by Ms. Emma He

More articles in Mathematics from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().