Võrdle meetodeid

	Mobile-Phone Mobility Estimation	Migrant Stock Estimation
Valdkond	Migration Studies	Migration Studies
Perekond	Process / pipeline	Process / pipeline
Tekkeaasta	2014	1983
Looja	Pierre Deville, Catherine Linard, Andrew J. Tatem, et al.	United Nations Population Division (standard measurement conventions)
Tüüp	Computational pipeline for population and migration inference from mobile data	Cross-source pipeline for counting the resident migrant population
Algallikas	Deville, P., Linard, C., Martin, S., Gilbert, M., Stevens, F. R., Gaughan, A. E., Blondel, V. D., & Tatem, A. J. (2014). Dynamic Population Mapping Using Mobile Phone Data. PNAS, 111(45), 15888-15893.	United Nations (1983). Manual on Methods of Estimating Internal Migration (Manual VI). Population Studies No. 47. New York: United Nations.
Rööpnimetused	CDR Mobility Estimation, Call-Detail-Record Migration Inference, Mobile Big Data Population Mapping, Phone-Based Displacement Tracking	Foreign-Born Stock Estimation, International Migrant Stock, Migrant Population Counting, Stock-Based Migration Measurement
Seotud	3	3
Kokkuvõte	Mobile-phone mobility estimation uses the digital traces left by ordinary phone use — call detail records, or CDRs — to map where people are, how their numbers shift over time, and how they move between places. Deville and colleagues' 2014 study in PNAS demonstrated that the locations of cell towers handling each call, aggregated across millions of subscribers, can produce dynamic population maps that track seasonal and daily changes far more finely than a decennial census ever could. Because a CDR records which tower served a user and when, the method can infer each person's habitual home location, count how many people 'live' in each area, and detect when those homes shift — the signature of internal migration or displacement. The approach turns a byproduct of telecom billing into a near-real-time demographic sensor, especially valuable where censuses are infrequent and crises move people faster than official statistics can follow. Crucially, the estimates are calibrated and validated against census or survey ground truth, so the phone-derived figures are anchored to known totals rather than taken at face value. The result is a powerful, if ethically fraught, way to observe human mobility at scale.	Migrant stock estimation answers a deceptively basic question: how many migrants are living in a place at a given moment? Unlike migration flows, which count moves over an interval, a stock is a cross-sectional count of people whose origin differs from their place of residence — most commonly the foreign-born, but sometimes the foreign-national or those who have lived abroad. The United Nations measurement conventions, set out in its migration manuals, fix the core definitions (place of birth versus citizenship, duration thresholds, usual residence) and the at-risk concepts that make stocks comparable. In practice the analyst rarely has one clean source: censuses give place-of-birth tables but miss recent or irregular arrivals, population registers give continuous citizenship-based counts but vary in how they handle departures, and surveys give detail but suffer sampling error. Migrant stock estimation is therefore a pipeline that compiles these sources, harmonizes their differing definitions and geographies, and adjusts for undercount, overstay, and double counting, drawing on the same comparability concerns Bell and colleagues raised for internal migration. The output — a coherent count of migrants by origin, age, and sex — underpins integration policy, flow estimation, and the denominators of countless migration indicators.