Given this sample of users, we construct a historical shadow profile for the location of each ego user based only on the data provided by its disclosing alters, as explained in the Data and Methods section.
Next, we evaluate the distribution of variance explained in our data as a function of class number as information loss is a concern.
Inclusion rules are explained in the data section.
For the same reason explained in "Simulated data for global graph clustering problem" section, it does not give good result if the cluster structure is weak.
When calculating the HHI and the CR3 we now only take competitors into account that are deemed relevant as explained in the data and methods section.
As explained in the data section, although the return intention dummy variable is available for 7,154 person-year observations for 827 immigrants in the sample, the intended migration duration variable is available for 6,099 person-year observations for 813 immigrants.
The methods for detecting the reaching onset and end are explained in the Data analysis section.
"You don't really see trends that are easily explained" in the data, Smith says.
As both types of correlation were relatively weak, we also performed a bootstrap analysis as explained in Supplementary Data S2.
