Migration component, both international and internal

BenjaminIsaac0111 / daedalus

Daedalus is a novel dynamic spatial microsimulation prototype being developed at the Leeds Institute for Institute Data Analytics. Currently, the main goal is to a create a proof of concept component as part of the SPENSER (Synthetic Population Estimation and Scenario Projection Model) project. In this we hope to demonstrate a potential for a highly integrated system that allows users to produce (custom) population projections for policy intervention analysis.

MIT License

4 stars 2 forks source link

Migration component, both international and internal #6

Closed kasra-hosseini closed 4 years ago

kasra-hosseini commented 4 years ago

[x] International migration as the first step. The effects of this parameter should be similar to mortality in the model.
[x] Input dynamic rates (currently, they are constant).
[x] It seems that, at some time steps, we remove some people twice (due to mortality rate and migration rate). This needs to be fixed (which should depend on the order of actions at one time step?)
[x] In the case of internal migration: how can we use the results in a spatial interaction model (allocating the internal migrants)? is it possible to calibrate the results using data (e.g., immigration rates from the census data)?

kasra-hosseini commented 4 years ago

A simplified version of the migration component is implemented here. One example output (considering both migration and mortality):

simple_example

Inputs:

time:
    start:
        year: 2011
        month: 1
        day: 1
    end:
        year: 2012
        month: 1
        day: 1
    step_size: 1  # Days
population:
    population_size: 0 # 1 Base
    age_start: 0
    age_end: 100
mortality:
    mortality_rate: 0.1
migration:
    migration_rate: 0.3

The sum of population/mortality/migration counts at the very last time step is 1024 which is higher than the initial population (1000 rows). My guess is that at some time steps, we remove some people twice (due to mortality rate and migration rate). This needs to be fixed (added to the TODO list).

The results can be reproduced in this notebook

kasra-hosseini commented 4 years ago

@crangelsmith Could you please confirm that this issue is resolved (see the first comment)? If yes, please go ahead and close it.

kasra-hosseini commented 4 years ago

Moved to: https://github.com/alan-turing-institute/daedalus/issues/14