AB - ES UID construction files

input_processing/00_master_conditions_ES.do

@@ -0,0 +1,186 @@
+/*******************************************************************************
+* PROJECT:             	SimPaths EU
+* DO-FILE NAME:        	00_master_conditions.do
+* DESCRIPTION:         	Sets out the assumptions and conditions imposed in the 
+* 						creation of the unique dataset and the if conditions 
+* 						imposed when estimating the processes for SimPaths.  
+********************************************************************************
+* COUNTRY:              ES
+* AUTHORS: 				Ashley Burdett
+* LAST UPDATE:          April 2026 AB
+********************************************************************************
+*   -----------------------------------------------------------------------
+*    Assumptions imposed to align the SILC data with simulation rules:
+*   -----------------------------------------------------------------------
+*
+*   - Retirement:
+*       - Treated as an absorbing state
+*       - Must retire by a specified maximum age
+*       - Cannot retire before a specified minimum age
+*
+*   - Education:
+*       - Leave education no earlier than a specified minimum age
+*       - Must leave the initial education spell by a specified maximum age
+*       - Cannot return to education after retirement
+*
+*   - Work:
+*       - Can work from a specified minimum age
+*       - Activity status and hours of work populated consistently:
+*           → Assume not working if report hours = 0 
+*           → Assume hours = 0 if not working
+* 		- If missing partial information, don't assume the missing is 0 and 
+* 			impute (hot-deck)
+*
+*   - Leaving the parental home:
+*       - Can leave from a specified minimum age
+* 		- Become the effective head of hh even when living with parents when 
+* 			paretns retire or reach state retirment age
+*
+*   - Home ownership:
+*       - Can own a home from a specified minimum age
+*
+*   - Partnership formation:
+*       - Can form a partnership from a specified minimum age
+*
+*   - Disability:
+*       - Treated as a subsample of the not-employed population
+*
+*   The relevant age thresholds are defined in globals defined in "DEFINE 
+* 	PARAMETERS" section below. 
+* 	Throughout also construct relevant flags and produce a log file 
+* 	"flag_descriptives.xlsx" to see the extent of the adjustments to the raw 
+* 	data. 
+*
+*   -----------------------------------------------------------------------
+*    Additional notes on implementation: 
+*   -----------------------------------------------------------------------
+*
+*   - Impute health score (generalized ordered logit model).
+*   - Constructing age is not straight forward as not directly reported in the 
+* 	  data, therefore: 
+*       → Use interview age (RX010) where available
+*       → Otherwise use age at end of interview year (PX020). This results in 
+* 			upward bias of age.
+*   - Set education = 0 (na) while in initial education spell. 
+*
+*   -----------------------------------------------------------------------
+*   Remaining disparities between initial populations and simulation rules:
+*   -----------------------------------------------------------------------
+*
+*   - Ages at which females can have a child. [Be informed by the sample?]
+*	  Permit teenage mothers in this script (deal with in 03_ )
+*   - A few higher/older education spells (30+) that last multiple years 
+*     in the simulation can only return to education for single year spells. 
+* 	- Number of children vars (all ages or 0-2) don't account for feasibility 
+* 		of age at birth of the mother. 
+*
+*******************************************************************************/
+
+
+/*******************************************************************************
+* DEFINTE PARAMETERS
+*******************************************************************************/
+
+global country "ES"
+
+global first_sim_year "2011"
+
+global last_sim_year "2024"
+
+
+* Define threshold ages
+/*
+The thresholds defined below ensure consistency between the assumptions 
+applied in the simulation and the structure of the initial population data. 
+They specify the ages at which certain life-course transitions are permitted 
+or enforced within the model. These limits reflect both modelling conventions 
+and empirical considerations drawn from observed data.
+*/
+
+* Age become an adult in various dimensions	
+global age_becomes_responsible 18 
+
+global age_becomes_semi_responsible 16 
+
+global age_seek_employment 16 
+
+global age_leave_school 16 
+
+global age_form_partnership 18 
+
+global age_have_child_min 18 	
+
+global age_leave_parental_home 18
+
+global age_max_leave_parental_home 45
+
+global age_own_home 18
+
+global age_can_retire 50
+
+global age_force_retire 75             
+
+global age_force_leave_spell1_edu 30   
+
+global age_have_child_max 49  	
+	// allow this to be informed by the data  
+
+global age_cannot_separate 79
+
+global age_max_dep_child 17   
+	// used for defining BU and number of dependent children variables in data construction   
+
+global age_adult 18 
+	// used when slicing up hh level income information in data construction
+
+/*******************************************************************************
+* PROCESS IF CONDITIONS 
+*******************************************************************************/
+
+* Education 
+global e1a_if_condition "dag >= ${age_leave_school} & dag < ${age_force_leave_spell1_edu} & l.les_c4 == 2 & flag_deceased != 1"
+
+global e1b_if_condition "dag >= ${age_leave_school} & l.les_c4 != 4 & l.les_c4 != 2 & flag_deceased != 1"
+
+global e2_if_condition "dag >= ${age_leave_school} & l.les_c4 == 2 & les_c4 != 2 & flag_deceased != 1"
+
+* Leave the parental home 
+global p1_if_condition "ded == 0 & dag >= ${age_leave_parental_home} & dag <= ${age_max_leave_parental_home} & flag_deceased != 1"
+
+* Partnership 
+global u1_if_condition "dag >= ${age_form_partnership} & ssscp != 1 & flag_deceased != 1"
+
+global u2_if_condition "dgn == 0 & dag >= ${age_form_partnership} & l.ssscp != 1 & dag < ${age_cannot_separate} & flag_deceased != 1 & flag_deceased != 1"
+
+* Fertility 
+global f1_if_condition "dag >= ${age_have_child_min} & dag <= ${age_have_child_max} & dgn == 0 & flag_deceased != 1"
+
+* Health 
+global h1_if_condition "dag >= ${age_becomes_semi_responsible} & flag_dhe_imp == 0 & flag_deceased != 1"
+
+global h2_if_condition "dag >= ${age_becomes_semi_responsible} & ded == 0 & les_c4 != 4 & les_c4 != 2 & flag_deceased != 1"
+
+* Home ownership
+global ho1_if_condition "dag >= ${age_own_home} & flag_deceased != 1"
+
+* Retirement 
+global r1a_if_condition "dcpst == 2 & dag >= ${age_can_retire} & flag_deceased != 1"
+
+global r1b_if_condition "ssscp != 1 & dcpst == 1 & dag >= ${age_can_retire} & flag_deceased != 1"
+
+* Wages
+global W1fa_if_condition "dgn == 0 & dag >= ${age_seek_employment} & dag <= ${age_force_retire} & deh_c4 != 0 & flag_deceased != 1"
+
+global W1ma_if_condition "dgn == 1 & dag >= ${age_seek_employment} & dag <= ${age_force_retire} & deh_c4 != 0 & flag_deceased != 1"
+
+global W1fb_if_condition "dgn == 0 & dag >= ${age_seek_employment} & dag <= ${age_force_retire} & deh_c4 != 0 & previouslyWorking == 1 & flag_deceased != 1"
+
+global W1mb_if_condition "dgn == 1 & dag >= ${age_seek_employment} & dag <= ${age_force_retire} & deh_c4 != 0 & previouslyWorking == 1 & flag_deceased != 1"
+
+* Capital income 
+global i1a_if_condition "dag >= ${age_becomes_semi_responsible} & flag_deceased != 1" 
+
+global i1b_if_condition "dag >= ${age_becomes_semi_responsible} & receives_ypncp == 1 & flag_deceased != 1" 
+
+* Labour supply
+//??

input_processing/data_construction/README_data_construction.txt

@@ -0,0 +1,15 @@
+Construction of the SimPaths Input Datasets 
+
+Date: 6/6/26
+Author: A Burdett
+
+Workflow summary: 
+
+1. Obtain the required longitudinal EU-SILC releases. 
+
+2. Run the GESIS compilation program using the earlier longitudinal release to produce the harmonised 2005-2020 D, H, R, and P datasets. 
+
+3. Run the extended version of the GESIS compilation program using the 2024 longitudinal release to produce the harmonised 2005-2023 D, H, R, and P datasets. (Compilation files contained in …./data_construction/SILC_construction/2005_2023_panel/.)
+
+4. Run the SimPaths UID construction files for the country of interest, contained in the ”data_construction/UID_construction/” folder. During this process, the variables affected by post-2020 coding changes are merged from the 2005-2020 dataset into the 2005-2023 dataset. This creates the UID file (“…ipop.dta”) which is used to both process estimation and for internal validation of SimPaths, as well as the annual cross-sections used to initialise the model. 
+

input_processing/data_construction/SILC_panel_construction/2005_2023_panel/do_files/00_master_construct_eusilc_panel.do

@@ -0,0 +1,112 @@
+********************************************************************************
+* PROJECT:             	SimPaths EU  
+* DO-FILE NAME:         00_master_construct_eusilc_panel
+* DESCRIPTION:          Imports the EU-SILC csv files. 
+********************************************************************************
+* COUNTRY:              -
+* DATA:         	    EU-SILC longitudinal data
+* AUTHORS: 				Ashley Burdett
+* LAST UPDATE:          Jan 2025
+********************************************************************************
+* NOTE:					This is a formatted, extended (to include 2023 data) 
+* 						and organized version of the procedure constrcuted by 
+* 						GESIS to combine the SILC longitudinal data panels in a 
+* 						systematic way to idenitfy individuals across 
+* 						observations. 
+* 
+* 						In addition to the included do-files you will need to 
+* 						obtain the set-up files that transform csv files 
+* 						containing the SILC data from GESIS. They can be found 
+* 						here:  https://www.gesis.org/en/missy/materials/EU-SILC/setups#:~:text=Routines%20to%20transform%20*.csv%20to,document%20of%20every%20data%20release.
+* 
+* 						The version of the set-up files used should coincide 
+* 						with the data release being use. 
+* 
+* 						Population files merged added in "05_weights" need to be
+* 						uploaded into the input_data subfolder. The 2005-2020 
+* 						information was obtained from GESIS, whilst the 
+* 						2021-2023 information was obtained from EUROSTAT. 
+* 		
+* 						Make sure the necessary subfolders are created for the 
+* 						set-up file outputs. 
+*						Currently called in do-file 01-04 using the following 
+* 						pattern
+* 						${do_dir}/EU-SILC/19 L-2023/ 
+* 						${do_dir}/EU-SILC/18 L-2022/
+* 						${do_dir}/EU-SILC/17 L-2021/ ...
+*
+********************************************************************************
+
+* Initalization 
+clear all
+capture log close
+set more off
+
+
+/*******************************************************************************
+* DEFINE DIRECTORIES
+*******************************************************************************/
+
+global main_dir "/Users/aburdett/Library/CloudStorage/Box-Box/ESPON - OVERLAP/_countries/Cumulative Longitudional Dataset (all countries)/2005_2023_panel" // master folder for processing
+
+global log_dir "$main_dir/logs" // folder in main_dir that contains logs
+
+global do_files_dir "$main_dir/do_files" // folder in main_dir that contains do files 
+
+global data_dir "$main_dir/data" // folder in main_dir that contains output data
+
+global input_data_dir "$main_dir/input_data"
+
+//for set-up files 
+global do_dir "$main_dir/data" // folder in main_dir that contains output data for merging do file 
+
+global datapath "$main_dir/data" // folder in main_dir that contains output data for merging do file 
+
+global csv_path "$main_dir/input_data"
+
+
+/*******************************************************************************
+* EXECUTE FILES
+*******************************************************************************/
+* Call set-up files to import the EU-SILC data 
+forvalues i = 2005/2023 {
+
+	do "$do_files_dir/Long_allcountries_Dfiles_`i'.do" 
+}	
+
+forvalues i = 2005/2023 {
+
+	do "$do_files_dir/Long_allcountries_Hfiles_`i'.do" 
+}	
+
+
+forvalues i = 2005/2023 {
+
+	do "$do_files_dir/Long_allcountries_Pfiles_`i'.do" 
+}	
+
+
+forvalues i = 2005/2023 {
+
+	do "$do_files_dir/Long_allcountries_Rfiles_`i'.do" 
+}	
+*/
+
+* Create panel using D files
+do "$do_files_dir/01_create_masterD.do" 
+
+
+* Create panel using H files
+do "$do_files_dir/02_create_masterH.do" 
+
+
+* Create panel using R files
+do "$do_files_dir/03_create_masterR.do" 
+
+
+* Create panel using P files
+do "$do_files_dir/04_create_masterP.do"
+
+
+* Adjust weight variables 
+do "$do_files_dir/05_weights.do"