Rolf Aaberge, Tarjei Havnes, and Magne Mogstad, "Ranking Intersecting
Distribution Functions", Journal of Applied Econometrics, Vol. 36, No.
6, 2021, pp. 639-662.

There are 2 folders with a total of 8 files. These are all zipped in
the file ahm-files.zip. All files are ASCII files in DOS format.
Unix/Linux users should use "unzip -a".

data folder
file		 description

eudist_hinc.csv  CSV file that contains the data for the first
		 application of the paper. Data come from the European
		 Community Household Panel (ECHP) for 1995–2001, and
		 from the European Union Statistics on Income and
		 Living Conditions (EU-SILC) for 2005–2010. Each
		 variable is labeled as hinc[country][year], where
		 country is the 2 letter abbreviation of the country.
		 Observation q is the observed q-th 1000-quantile of
		 the empirical distribution of individual equivalent
		 income, after adjusting for inflation and differences
		 in household size and composition, restricted to the
		 sample to households with a male aged 25-64.

jobsfirst.csv   CSV file that contains the data for the second
		application of the paper. Data come from the
		Connecticut’s Jobs First experiment. For detailed
		information about the program, refer to Bloom et al.
		(2002) or Bitler et al. (2006). The data contain the
		estimated quantile treatment effects estimated by
		Bitler et al. (2006). The file contains the following
		variables: quantile (the percentile), qte (the
		etsimated quantile treatment effect), level_comp
		(percentile of the income distribution of the
		comparison group), level_treat (percentile of the
		income distribution of the treatment group),
		treat_flattax (same as level_treat but where the
		average cost of the treatment per unit is subtracted
		from each percentile), treat_proptax (same as
		level_treat but where the cost of the treatment per
		unit is subtracted from each percentile proportional
		to their share of total income), qteflattax (=
		treat_flattax - level_comp), and qteproptax (=
		treat_proptax - level_comp)

code folder
file               description

anuk1994to2001.do  Do-file producing the results for the first
		   application of the paper, which uses UK data to to
		   study how the distribution of household income
		   evolved over a boom and a bust era in the British
		   economy.

anjobsfirst.do     Do-file producing the analysis for the second
		   application of the paper, which uses
		   random-assignment data to evaluate the
		   distributional effects of Connecticut’s Jobs First
		   program.

dominance.ado      Main estimation program to estimates the upwards and
                   downwards inverse stochastic dominance

multisort.ado	   Auxiliary function used in main estimation program

myfaculty.ado	   Auxiliary function used in main estimation program

myinteg.ado	   Auxiliary function used in main estimation program


The main estimation codes (anuk1994to2001.do and anjobsfirst.do)
export all the output in a subfolder that should be created inside the
replications folder. This folder should be called "logs". The user can
modify this behavior by changing this in the codes directly

Magne Mogstad <magne.mogstad [AT] gmail.com>