Latest results

1. Re-study of regional commuting in Estonia.
(R. Ahas, S. Silm, 2013)
The results of the study of commuting made on the basis of mobile positioning data shows that quantitatively it is possible to distinguish more than 50 pull centers in Estonia. Strength of the flows are shown by the number of commuters and the extent of the hinterland linked to the center.
 Regionaalne pendelränne
Centers on this figure are the main destinations for at least three residential areas.
2. Spatial Mobility between Tallinn and Helsinki in Mobile Positioning Datasets. Statistical overview.
(S. Silm, R. Ahas, M. Tiru, 2012)
This study report provides an overview of people’s mobility between Estonia and Finland with focus on the routes Tallinn–Helsinki and Helsinki–Tallinn. The traffic between the capitals of the two neighbouring countries is heavy and with millions of trips per year, it is an important tourism link and trade route. The capitals are also in active cooperation in the fields of business, administration and culture, which is the reason why they could be considered twin cities according to many parameters. Yet, the meaning of “twin cities” is complex, and this report does not aspire to evaluate the connections between Helsinki and Tallinn.
The objective of this study is to provide an overview of people’s mobility between the two countries and two cities.
Study results here:
3. Point pattern spatial interpolation method for mobile positioning data.
(E. Saluveer, L. Murov, A. Aasa, 2012)
Using mobile network cellular data brings up challenges on how to spatially interpolate the positioning results. Main interpolation methods used on cellular data are isopleth maps, centroid containment and area weighted interpolation of voronoi tesselation. All of these methods make a presumption of continuous surfaces, but if our main target is to map human whereabouts, we must not interpolate them in all directions continuously.
Point pattern analysis is widely used in ecology on spatial interpolation of species. Our study used point pattern generation using dasymetric mapping technique, where each positioning result was given probabilistic location according to characteristics of positioning point and land use. We tested this methodology on passive mobile positioning data from Estonian mobile operator EMT.
Our result show that point pattern generation methodology is promising as accuracy of positioning results aggregated to the municipality level were much better than with traditional methodology. Accuracy of positioning increased foremost in hinterland and rural areas.
4. Ethnic segregation out of home and work district.
(S. Silm, R. Ahas, E. Saluveer, 2012)


5. Final study report of regional commuting in Estonia.
(R. Ahas, S. Silm, K. Leetmaa, T. Tammaru, E. Saluveer, O. Järv, A. Aasa, M. Tiru, 2010)
Have a look on the study here: