Martin Huber, Lukas Laffers, and Giovanni Mellace, "Sharp IV Bounds on
Average Treatment Effects on the Treated and Other Populations under
Endogeneity and Noncompliance", Journal of Applied Econometrics,
Vol. 32, No. 1, 2017, pp. 56-79.

The ASCII file Colombia_data.txt contains the data used to create
Tables 6 and 7. The data contain 1201 observations and 4 variables and
were taken from the publicly available dataset of

  Angrist, J., Bettinger, E., Bloom, E., King, E. and Kremer, M.,
  2002, "Vouchers for private schooling in Colombia: Evidence from a
  randomized natural experiment", The American Economic Review 92,
  1535-1558.

Maximum grade finished (Y table 7)
Won a voucher (Z)
Going to private school (D)
Never repeating a grade (Y table 6)

The file Colombia_data.txt is zipped in the file hlm-data.zip.
Unix/Linux users should use "unzip -a".

To replicate Tables 6 and 7, please use the matlab code provided in
the file hlm-tabs-6-and-7-matlab.zip

The file BoundsLinprog.zip contains a matlab script which:

-- Checks that identified set for the share of defiers is an interval
for the dataset used in the paper

-- Compares the Analytical bounds derived in the paper with the
LinProg bounds

-- Mechanically checks the construction of the proof of the lemma that
states that the identified set for the share of defiers is an
interval.

-- Mechanically checks the construction of the proof of the sharpness
of the bounds within principal strata.

More instructions are provided in the file read_me.txt included in
BoundsLinprog.zip.

The file Colombia_data.dta contains the same data in Stata format. It
is zipped in the file hlm-stata-data.zip.

Giovanni Mellace
giome [AT] sam.sdu.dk