An introduction to spatial microsimulation with R

These are the slides to accompany a two day course with the same name. See https://github.com/Robinlovelace/smsim-course for more information.

Transcript

1. An introduction to spatial
   microsimulation with R
   A Leeds Short Course funded by the CDRC
   Robin Lovelace, University of Leeds
   See http://robinlovelace.net/smsim-course/
   28th May 2015
   

   View Slide

2. About this course
   ●
   Aims to introduce both theory and practice of
   spatial microsimulation
   ●
   Divided into 3 main parts:
   – Foundations
   – Creating spatial microdata with R
   – Applying the method
   ●
   Housekeeping
   

   View Slide

3. The agenda... and the handout
   ●
   9:30 – 11:00 Introduction
   – What is it? What it does.
   Reweighting
   ●
   11:15 – 13:00 Smsim in R
   – Loading the data, IPF in R
   ●
   13:30 – 16:30 An example
   – Preparing input data
   ●
   9:30 – 11:00 Analysis
   – Validation
   ●
   11:15 – 13:30 Extensions
   – Integerisation
   1) Foundations
   2) IPF in R
   3) CakeMap /
   other
   

   View Slide

4. Why do you need spatial
   microsimulation? I
   ●
   To tackle the modifiable areal unit problem
   (MAUP) (Openshaw 1983)
   The world is like this: complex
   Administrative zones
   (“areal units”) look like this
   - a little arbitrary!
   Making it
   look (a bit)
   like this:
   oversimple!
   

   View Slide

5. Why do you need spatial
   microsimulation? II
   As an input into
   agent based
   models (image is
   in the context of
   transport
   modelling –
   deterministic
   reweighting =
   spatial
   microsimulation)
   To create a
   synthetic
   population, also
   known as synthetic
   spatial microdata
   

   View Slide

6. What is spatial
   microsimulation? I
   ●
   A method for
   synthesising spatial
   microdata based on
   survey and areal input
   data (Lovelace and
   Ballas 2013)
   

   View Slide

7. What is spatial microsimulation? II
   ●
   A procedure to translate
   from 'wide' to 'long' data
   formats
   ●
   Restrictive data anonymity
   regulations
   

   View Slide

8. Applications – some examples
   ●
   To estimate local smoking rates (Tomintz et al.
   2008)
   ●
   To investigate commuter patterns and model
   distributional impact of future scenarios
   (Lovelace et al. 2014)
   ●
   Farmer participation in agri-environment
   schemes (Hynes et al. 2008)
   

   View Slide

9. Microsimulation and ABMs
   

   View Slide

10. Validation
    ●
    It's tricky – you'll generally only simulate things
    you do not know
    ●
    But very important: possible to mislead with this
    technique
    ●
    Internal validation: compare with prior
    expectations – is the model working OK?
    ●
    External validation: compare with the real world
    

    View Slide

11. References
    ●
    Edwards, K. L., Clarke, G. P., Thomas, J., & Forman, D. (2011). Internal and external
    validation of spatial microsimulation models: Small area estimates of adult obesity.
    Applied Spatial Analysis and Policy, 4(4), 281-300. (The importance of validation).
    ●
    Hynes, S., Farrelly, N., Murphy, E., & O'Donoghue, C. (2008). Modelling habitat
    conservation and participation in agri-environmental schemes: a spatial microsimulation
    approach. Ecological economics, 66(2), 258-269. (Agricultural application).
    ●
    Lovelace, R., Ballas, D., & Watson, M. (2013). A spatial microsimulation approach for the
    analysis of commuter patterns: from individual to regional levels. Journal of Transport
    Geography. (Policy-relevant application).
    ●
    Lovelace, Robin, and Dimitris Ballas. ‘Truncate, replicate, sample’: A method for creating
    integer weights for spatial microsimulation. Computers, Environment and Urban Systems
    41 (2013): 1-11. (Method).
    ●
    Openshaw, S. (1983). The modifiable areal unit problem (Vol. 38). Norwich: Geo Books.
    ●
    Tomintz, M. N., Clarke, G. P., & Rigby, J. E. (2008). The geography of smoking in Leeds:
    estimating individual smoking rates and the implications for the location of stop smoking
    services. Area, 40(3), 341-353. (Health application).
    

    View Slide