Abstract
Alternative data allow researchers and policy makers to measure the real economic activities at high frequencies and resolutions official statistics cannot match. The gains are largest in lower-middle-income economies, where official series arrive with long lags or are of limited coverage, mobile networks reach further than conventional statistical infrastructure. This paper builds such a proxy for it from anonymized mobile-network subscriber’s location information supplied by the Cambodian mobile-network operator (Smart Axiata). The pipeline identifies factory sites from the MISTI registry, matches each to a nearby cell coverage area, and classifies subscribers with persistent weekday daytime presence as workers, yielding a balanced panel of 38 sites across 17 provinces over January 2025 – June 2026. The headline index, post-stratified to the national distribution of big factories, tracks the seasonal production cycle and co-moves with official manufacturing exports. A difference-in-differences design around the 2025 Cambodia–Thailand border tensions shows large negative effects at border-province SEZs and smaller effects at individual factories. Descriptive subgroup estimates suggest the burden fell most heavily on the youngest workers and that the post-December recovery at individual factories was faster for male than for female workers.
Keywords: Alternative Data; Mobility Data; Nowcasting; Manufacturing; Cambodia; Industrial Production; Mobile Phone Records.
About the presenter
Join us at the Forum, 5–6 November 2026, Institute of Technology of Cambodia.
Are you the author? Sign in to My NRF to share this abstract.