Nowcasting Automobile Sales in Pakistan: An Empirical Assessment Using Google Trends
This study attempts to nowcast car sales in Pakistan's automobile sector by incorporating a Google Trends Automobile Index (GTAI) into a traditional nowcasting framework. By employing secondary data over the period of July 2010 to June 2024, this study evaluates nowcasting performance by modeling a Car Sales Index (CSI) constructed from the Industrial Production Index (IPI) and monthly sales of five selected car brands against a Google Trends Automobile Index (GTAI) derived from the popularity of key search words on Google. Using ARMA Maximum Likelihood estimation, the study empirically examines whether integrating a GTAI into a traditional ARMA model improves monthly car sales nowcasting. The empirical results indicate that while the GTAI predicts car sales in univariate models, it becomes redundant once the IPI is included. This statistical ineffectiveness in Nowcasting is driven by the late surge in broadband internet penetration and the predominantly young age of internet users, who typically do not influence household car-purchasing decisions. These dynamics suggest that, future nowcasting frameworks may refine their underlying search terms and incorporate localized platforms like PakWheels.com to achieve more accurate predictions.
Authors
- Hafiz Abdur Rehman (ORCID: https://orcid.org/0000-0003-4882-6947)
- Tauqir Ahmed
- Arshad Ali Bhatti (ORCID: https://orcid.org/0000-0002-5752-6623)
- Atiq Hassan Shah
Institutions
- International Islamic University, Islamabad (PK)
Publication Details
- Journal
- Review of Economics and Development Studies
- Published
- 2026-11-20
- DOI
- https://doi.org/10.47067/reads.v12i4.603
- Primary Topic
- Forecasting Techniques and Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00