Konstantinos Angelopoulos, Spyridon Lazarakis, and James Malley,
"Cyclical Labour Income Risk in Great Britain", Journal of Applied
Econometrics, Vol. 37. No. 1, 2022, pp. 116-130.

The zip file alm-files.zip contains all of the files necessary to
replicate the results in this paper. Please contact Spyridon Lazarakis
if you have any questions regarding the codes (email address:
s.lazarakis [AT] lancaster.ac.uk).

Software

The research was conducted using STATA version 14 and Matlab version 2018a.

Data

Due to data protection regulations, we are not permitted to distribute
the data.  However, the data are available for scientific research
free of charge from the UK Data Service (see, ukdataservice.ac.uk). 
Accessing the data requires each researcher to register with the
Service.

1. Bardasi, E., Jenkins, S.P., Sutherland, H., Levy, H., Zantomio, F.
   (2012). British Household Panel Survey Derived Current and Annual
   Net Household Income Variables, Waves 1-18, 1991-2009. [data
   collection]. 9th Edition. UK Data Service. SN: 3909,
   doi.org/10.5255/UKDA -SN-3909-2.

2. University of Essex, Institute for Social and Economic Research.
   (2018). British Household Panel Survey: Waves 1-18, 1991-2009.
   [data collection].  8th Edition. UK Data Service. SN: 5151,
   doi.org/10.5255/UKDA -SN-5151-2.

Instructions

1. Copy the following datasets: (i) UKDA-3909-stata10; and (ii)
   UKDA-5151-stata into the folder named
   \Data_Code_Risk_UK_JAE_July_2021\raw_data. It is important that the
   folder structure is not changed for the program to work without
   error.

2. Insert the directory name of your main folder into all of the Stata
   .do files.

3. Run the main file: \Data_Code_Risk_UK_JAE_July_2021\0_main_
   file_BHPS.do. This will process all the Stata files relating to the
   BHPS data. The do file provides comments on the different steps of
   the data cleaning and how the relevant data is extracted. See the
   paper for further details on the sample selection.

4. Run the main file A_main_bootstrap_file.m in each of the following
   folders: i) Section4.1; ii) Section4.2_periods91; iii)
   Section4.2_periods97; iv) Section5; and v) SectionAppendixE . Each
   job requires approximately 24 hours to run on a computer with
   Matlab2018a and 12 physical cores.