Integration of MNO data with survey data to produce commuting statistics
Conference
65th ISI World Statistics Congress 2025
Format: IPS Abstract - WSC 2025
Keywords: data-integration, mobile phone, official-statistics
Session: IPS 934 - Integrating Mobile Network Operator Data with Official Statistics
Thursday 9 October 2 p.m. - 3:40 p.m. (Europe/Amsterdam)
Abstract
In this work different approaches for integrating Mobile network operator (MNO)
data with survey data to produce official statistics are explored.
The study is developed under the MNO MINDS ESSnet.
Traditionally, decennial population censuses have been the basis for gathering commuter data, proving flows of individuals across municipalities. However, for more timely information, some NSIs have shifted towards annual census surveys with reduced sample sizes. This transition presents various challenges in deriving accurate commuting statistics.
Availability of MNO data allows for the derivation of recurrent flows between
home and work/study locations. These flows serve as a valuable information
and can be used as the primary source for producing official statistics by adjusting MNO data coverage with a quasi randomisation approach.
In the superpopulation framework, aggregate MNO data serve as covariates,
complementing target variables obtained from census sample surveys.
Spatial interaction models are widely used in transportation planning and
urban studies to describe the movement of people, goods or information across
different places.
These models describe the relationship between the origin and destination
of flows within a geographic area accounting for factors such as distance, attractiveness, and connectivity of the different locations, also including spatial
auto-regressive components.
To reduce bias that may arise due to miss-specification of the previous models, small area models can be applied.
The methods are applied to estimate commuter flows
between municipalities of an Italian region.