Dynamic Micro-Simulation for Population Projection - An Application to Mauritania

The DYNAMIS-POP-MRT_v01_2017 model

• Dynamic Micro-Simulation for Population Projections - An Application to Mauritania
• Acknowledgments
• Abstract
• Chapter 1. Rationale, Methods, Implementation, and a Portable Application
  • 1.1. Introduction: Purpose, Scope, Organization
    • References
  • 1.2. The Dynamic Micro-simulation Approach
    • 1.2.1. Advantages
      • Population heterogeneity
      • The problem of aggregation
      • Individual and linked histories
    • 1.2.2. Drawbacks
      • Detail versus prediction power
      • Transitory limitations
    • 1.2.3. Types of dynamic micro-simulation models
    • References
  • 1.3. Population Projections by Micro-simulation: Rationale
    • References
  • 1.4. A Portable Modular Application
    • 1.4.1. Starting population
    • 1.4.2. Fertility
      • The Base Fertility Module
      • The Refined Fertility Module
    • 1.4.3. Mortality
      • The Base Mortality Module
      • The Optional Child Mortality Module
    • 1.4.4. Internal Migration
      • The Base Internal Migration Module
      • The Refined Internal Migration Module
    • 1.4.5. Immigration
    • 1.4.6.Emigration
    • 1.4.7. The Primary Education Module
    • 1.4.8. First Union Formation Module
    • 1.4.9. Running the Model
    • 1.4.10. Model Output
  • 1.5. Data Requirements
    • 1.5.1. Overview
    • 1.5.2. Data Files
      • Population and Housing Census 2013
      • Multiple Indicator Cluster Survey (MICS) 2011
• Chapter 2. Phase 1: Reproducing a Typical Macro Population Projection Model
  • 2.1. The Base model: Reproducing a macro model
  • 2.2. Starting Population
    • 2.2.1. Discussion
    • 2.2.2. Parameters
    • 2.2.3.Analysis
  • 2.3. Fertility
    • 2.3.1. Discussion
    • 2.3.2. Parameters
    • 2.3.3. Analysis
  • 2.4. Mortality
    • 2.4.1. Discussion
    • 2.4.2. Parameters
    • 2.4.3. Analysis
  • 2.5. Internal Migration
    • 2.5.1. Discussion
    • 2.5.2. Parameters
    • 2.5.3. Analysis: Estimating the Parameters
  • 2.6. Immigration
    • 2.6.1. Discussion
    • 2.6.2. Parameters
    • 2.6.3. Analysis
  • 2.7. Emigration
    • 2.7.1. Discussion
    • 2.7.2. Parameters
    • 2.7.3. Analysis
• Chapter 3. Phase 2: Extending the Model Showcasing the Power of Micro-simulation
  • 3.1. Primary Education
    • 3.1.1. Discussion
    • 3.1.2. Parameters
    • 3.1.3. Analysis
      • The core models
      • Introducing Mother’s education
      • Reproducing the model
  • 3.2. Union Formation
    • 3.2.1. Discussion
    • 3.2.2. Parameters
    • 3.2.3. Analysis
      • Current period data
      • Hazard regression, including age-specific trends
      • Hazard regression by cohort
      • Parametric models
      • Reproducing the module
  • 3.3. Alternative Models for Fertility
    • 3.3.1. Discussion
    • 3.3.2. Parameters
    • 3.3.3. Analysis
      • First births
      • Higher-order births
      • Reproducing the analysis
  • 3.4. Infant Mortality
    • 3.4.1. Discussion
    • 3.4.2. Parameters
    • 3.4.3. Analysis
  • 3.5. Migration refinements
    • 3.5.1. Discussion
    • 3.5.2. Parameters
    • 3.5.3. Analysis
    • 3.5.4. Reproducing the Analysis
• Chapter 4. Using the model
  • 4.1. Analysis Example: Demographic Effects of Education
    • 4.1.1. Scenarios
    • 4.1.2. Results
  • 4.2. Analysis Example: De-Composing Changes in Child Mortality
    • 4.2.1. Scenarios
    • 4.2.2. Results
  • 4.3. Analysis Example: Internal Migration by Education
    • 4.3.1. Scenarios
    • 4.3.2. Results
      • Zero Migration Scenario A0
      • The Base Scenario A: Continuing Educational Trends, Base Migration (Without Education)
      • Scenario B: Universal Primary Education, Base Migration (Without Education)
      • Scenario C: Continuing Educational Trends, Refined Migration (With Education)
      • Scenario D: Universal Primary Education, Refined Migration (With Education)
• Chapter 5. Micro-simulation Technology: Modgen/OpenM++
  • 5.1. The User’s Perspective: User Interface
    • 5.1.1. Running a model
    • 5.1.2. Scenario Settings
    • 5.1.3. Model Results
    • 5.1.4. Output Table Views
    • 5.1.5. Exporting Simulation Results
    • 5.1.6. Further documentation
  • 5.2. The Developer’s Perspective: Key Programming Concepts
    • 5.2.1. Parameter Dimensions and Tables
    • 5.2.2. Actors, states, functions and event declarations
    • 5.2.3. Event Implementation
    • 5.2.4. Table output
    • 5.2.5. Model documentation
    • 5.2.6. Programming Wizards and Model Templates
• Chapter 6. Statistical Measures, Concepts, and Methods Common in Micro-simulation
  • 6.1. Starting a Simulation and Keeping it Going: Population, States, Events
  • 6.2. Measures of the likelihood of events: probabilities and rates
  • 6.3. From Measures to a Simulation
  • 6.4. Estimating and calibrating probabilities and rates
  • 6.5. Probability Distributions and Origin-destination Matrices
  • 6.6. Micro-simulation Implementation
• Chapter 7. A Step-by-Step Guide
  • 7.1. Step 1: Model Templates
    • 7.1.1. Overview
      • Model Templates
      • Model Versions
    • 7.1.2. Concepts
      • Overview
      • Code organization in modules
      • Code organization within modules
      • Documentation:
      • Type definitions
      • Parameters
      • Declarations of actor states, functions, and events
      • Implementation of events
      • Uniform distributed random numbers
      • Special Functions: Start() and Finish()
      • Tables
    • 7.1.3. How to reproduce this modeling step
      • Useful Visual Studio settings:
    • 7.1.4. Model versions
      • Converting the model to a template for time-based models
      • The module ModelSettings.mpp in a time-based version at Step 0
  • 7.2. Step 2: Creating a Population
    • 7.2.1. Overview
    • 7.2.2. Concepts
      • Overview
      • Adding a new module
      • Sampling from a distributional parameter table
    • 7.2.3. How to reproduce this modeling step
      • Adding new types in PersonCore.mpp
      • New states in PersonCore.mpp
      • Create a new module StartPopSampling.mpp
      • Code of the new module StartPopSampling.mpp
      • Modify the Start() function in PersonCore.mpp
      • Modify timeMortailityEvent() in Mortality.mpp
      • Adding the parameter and values in the Base(PersonCore).dat file
      • Add and modify tables in Tables.mpp
    • 7.2.4. Model versions
      • Converting the model to a time-based and/or micro-data-based version
    • 7.2.5. The module ModelSettings.mpp in alternative versions of Step 1
      • Sampling - time based
      • Micro-data file - case based
      • Micro-data file - time based
  • 7.3. Step 3: Population Scaling
    • 7.3.1. Overview
    • 7.3.2. Concepts
      • How to scale model output
      • Scaling in case-based models
      • Scaling in time-based models
    • 7.3.3. How to reproduce this step
      • Changes in StartPopSampling.mpp
      • Code in the module ModelSettings.mpp
      • Changes in PersonCore.mpp
    • 7.3.4. Model versions
      • Converting the model to a time-based and/or micro-data based version
      • The module ModelSettings.mpp in alternative versions of Step 2
  • 7.4. Step 4: Fertility
    • 7.4.1. Overview
    • 7.4.2. Concepts
      • Overview
      • RANGE_POS
      • Model-generated parameters
      • Pre-Simulation
      • Creating and actor and inheriting characteristics
    • 7.4.3. How to reproduce this modeling step
      • The module Fertility.mpp
      • Modification of the Start() Function in PersonCore.mpp
      • Add parameters to the data file and initialize them with values
      • Tables
      • Modifications of an existing table
    • 7.4.4. Model versions
      • Converting the model to a time-based and/or micro-data-based version
  • 7.5. Step 5: Refining Mortality
    • 7.5.1. Overview
    • 7.5.2. Concepts
      • Example for a binary search
    • 7.5.3. How to reproduce this modeling step
      • Code changes and new code in Mortality.mpp
    • 7.5.4 Model versions
      • Converting the model to a time-based and/or micro-data-based version
  • 7.6. Step 6: Migration
    • 7.6.1. Overview
    • 7.6.2. Concepts
      • Transition matrices
      • SPLIT(), split(), self_scheduling_split()
      • Converting probabilities to rates
    • 7.6.3. How to reproduce this modeling step
      • The module Migration.mpp
      • New tables in Tables.mpp
    • 7.6.4. Model versions
      • Converting the model to a time-based and/or micro-data-based version
  • 7.7. Step 7: Emigration
    • 7.7.1. Overview
    • 7.7.2. Concepts
    • 7.7.3. How to reproduce this modeling step
      • The Emigration.mpp module
      • Parameters
      • Actor definitions
      • Event implementation
      • Tables
    • 7.7.4. Model versions
      • Converting the model to a time-based and/or micro-data-based version
  • 7.8. Step 8: Immigration
    • 7.8.1. Overview
    • 7.8.2. Concepts
      • Copying parameters from double to cumrate tables in pre-simulation
      • Population scaling
    • 7.8.3. How to reproduce this modeling step
      • The Immigration.mpp module
      • Changes in PersonCore.mpp
      • Changes in ModelSettings.mpp
      • Changes in the Start() function - 3 types of persons and population scaling
      • New tables
    • 7.8.4. Model versions
      • Converting the model to a time-based and/or micro-data-based version
      • The module ModelSettings.mpp in alternative versions of Step 7
  • 7.9. Step 9: Micro-Data Output
    • 7.9.1. Overview
    • 7.9.2. Concepts
      • Modgen functionality for micro-data output
    • 7.9.3. How to reproduce this modeling step
      • The MicroDataOutput.mpp module
    • 7.9.4. Model versions
      • Converting the model to a time-based and/or micro-data-based version
  • 7.10. Step 10: Micro-Data Input
    • 7.10.1. Overview
    • 7.10.2. Concepts
      • Modgen functionality for micro-data input
    • 7.10.3. How to reproduce this modeling step
      • The StartPopFile.mpp module
      • New settings in ModelSettings.mpp
      • The microdata file
  • 7.11. Step 11: Conversion to a Time-Based Model
    • 7.11.1. Overview
    • 7.11.2. Concepts
      • Time-Based Models
    • 7.11.3. How to reproduce this modeling step
      • New settings in ModelSettings.mpp
    • 7.11.4. Discussion: Model type conversions
      • The DYNAMIS-POP-MRT.mpp file for time-based models (all steps)
      • The StartPopFile.mpp module for reading in a starting file
  • 7.12. Step 12: Sampling from a detailed starting population
    • 7.12.1. Overview
    • 7.12.2. Concepts
      • Adding a new actor type
      • Actor Sets
    • 7.12.3. How to reproduce this modeling step
      • Changes to file operations for allowing multi-threading
      • Input file variables and file lenght (StartPopFile.mpp)
      • The new actor type Observation (StartPopFile.mpp)
      • Using the new Observation actor for sampling starting characteristics (StartPopFile.mpp)
      • Changes in the Simulation() function (DYNAMIS-POP-MRT.mpp)
      • Changes in the Start() function
  • 7.13. Step 13: Primary Education
    • 7.13.1. Overview
    • 7.13.2. Concepts
      • Gaining efficiency by avoiding large numbers of individual level events happening at the same time
      • Sampling characteristics from a donor population
      • Aligning model results
    • 7.13.3. How to reproduce this modeling step
      • The PrimaryEducation.mpp module
      • The actor Primary School definitions and basics: Start, Finish, School year change
      • Initialization of the primary education status in the starting population
      • Primary school entry
      • Primary school graduation
      • Primary education status of immigrants
      • Supporting functions
    • 7.13.4. Table Output
      • Changes in the Simulation() function in DYNAMIS-POP-MRT.mpp
      • Changes in PersonCore.mpp
      • Changes in SatrtPopFile.mpp
      • Changes in the Immigration.mpp module
  • 7.14. Step 14: First Union Formation
    • 7.14.1. Overview
    • 7.14.2. Concepts
      • Allowing users to choose between models
    • 7.14.3. How to reproduce this modeling step
      • The UnionFormation.mpp module
      • Calculating hazards using a Coale McNeil model
      • The union formation event
      • Sampling the union formation age for immigrants
      • Presimulation
      • Tables
      • Changes in SetAliveEvent()
      • Changes in GetStartCharacteristics()
  • 7.15. Step 15: Fertility refined
    • 7.15.1. Overview
    • 7.15.2. Concepts
      • State duration: Time intervals since an event
      • Model Alignment: Using one model for scheduling events happening to persons selected by another model
    • 7.15.3. How to reproduce this modeling step
      • The new Fertility.mpp module
      • Changes in StartPopFile.mpp
      • Changes in UnionFormation.mpp to assign values to immigrants
  • 7.16. Step 16: Infant mortality
    • 7.16.1. Overview
    • 7.16.2. Concepts
      • Calibrating a model during execution
      • Actor linkages
    • 7.16.3. How to reproduce this step
      • The new module ChildMortality.mpp
      • Changes in the simulation engine DYNAMIS-POP-MRT.mpp
      • Changes in the general mortality module Mortality.mpp
    • 7.16.4. Changes in the PersonCore.mpp module
  • 7.17. Step 17: Migration refined
    • 7.17.1. Overview
    • 7.17.2. Concepts
    • 7.17.3. How to reproduce this step
  • 7.18. Step 18: Extending Table and Micro-Data Output
    • 7.18.1. Overview
    • 7.18.2. Concepts
      • Adding a header line with variable names to csv files
    • 7.18.3. How to reproduce this step
      • Extending the Micro-Data Output Options
      • The output event
      • Tables
      • Table groups
      • States and dimensions used in tables
      • General Population Tables
      • Fertility Tables
      • Mortality Tables
      • Migration Tables
      • Education Tables
      • Union Formation Tables
      • Female Life-Course Experience Tables
• Chapter 8. Software Downloads
  • 8.1. Model Users
    • 8.1.1. Prerequisites
    • 8.1.2. DYNAMIS-POP-MRT Phase 1 (User Version)
    • 8.1.3. DYNAMIS-POP-MRT Phase 2 (User Version)
  • 8.2. Model Developers
    • 8.2.1. Required Software for Modgen Developers
    • 8.2.2. The DYNAMIS-POP-MRT model (full developer’s version)