• Andy Turner's MoSeS Technical Meeting 2007-12-11 Web Page

  • [An image of Andy Turner]

  • Introduction

    • Web Page about the MoSeS Technical Meeting on 2007-12-11.
    • Contents
      • People
      • Documentation
      • Notes
      • References
      • Validation and Metadata

  • People

    • Mark Birkin (Chair)
    • Paul Townend
    • Andy Turner
    • Belinda Wu

  • Documentation

  • Notes

    • Meeting Focus
      • Architecture outside of the portal and pulling all the modules together.
    • File/database structure and data formats.
      • Metadata
      • XML Schema
        • Paul thinks that using XML might make things too slow and inefficient.
        • Schemas do provide a means (the only?) of having a flexible datastore:
          • The schema is a mapping of variable names to URI's.
          • The URI's are mappings of a MoSeS ID to a value.
          • A MoSeS ID is a unique Household or Individual ID.
        • The data store needs to be extensible in that new attributes might be linked at any time.
          • To do this all one would need to do is create a new schema (inheriting an existing schema and adding an extra variable).
        • Schemas are not to be nested (a URI mapped value should not point to another Schema).
    • Mark has developed a bastardised version of the dynamic demographic model.
    • Modularisation
      • Want to be able to plug and play, so the dynamic model, that is currently inputting Mark's Population Initialisation data, is able to input Andy's Population Initialisation data and data from BHPS provided by Chengchao's linking work.
    • Focus on the Dynamic Model
      • Modules:
        1. Mortality
          • Data Inputs:
            • From Population Initialisation data; Age, Location , Sex (Gender): From some ONS Ward Level data; probability
          • Process:
            1. Random probabilistic death applied individually (Unconstrained at an area level).
            2. _IsAlive status variable changed for those who die at 1.
        2. Fertility
        3. Health Status
        4. Marriage
          • Data Inputs:
            • ?
          • Process:
            1. Select Not Married, age > 16
            2. Sort by Sex (Gender)
            3. Match (by age assignment)
            4. Do they move?
          • Mention of a Pool Mode?
        5. Migration
        6. Ageing
    • Data Transfer and Repository
      • Let's look to use the new NCeSS portal, but we could use BSCW for now.
    • Actions
      • Andy and Paul to write up and distribute their notes.
      • Andy, Paul and Belinda to discuss each module of her Dynamic Model.
        • Andy to generate a set of input Schemas for the data inputs to each Dynamic Model Module.
        • Andy to produce a list of what variables are updated by each Dynamic Model Module.
      • Andy to work out how best to transfer the data for us to share now and what to use with regard a common repository for code.
        • This is not the portal archive or SRB instance that we might use for data processing on a live system.
      • Belinda to provide Andy and Paul with data and latest version of Dynamic Model code so they can run it.
      • Andy and Paul to look at Belinda's code and start thinking about optimisation.
      • Mark to organise next technical meeting.

  • References

    • Andy Turner's MoSeS Meeting Web Page

  • Validation and Metadata

      [Validate CSS]
    • Version 0.1.0 of this page published on 2008-01-04.
    • Page hosted on the School of Geography webserver at the University of Leeds.
    • Copyright: Andy Turner, University of Leeds