diffA

a design matrix that evaluates the MTRs over the grid
generated for the audit procedure.

the grid generated for the audit procedure, sorted
by the values of the covariates, and monotone increasing in the
unobservable.

monogrid

the number of binding constraints in the S-set.

indicator for treatment group (<code>d=1</code>) versus control
group (<code>d = 0</code>).

number of terms in the MTR for the other experimental
group. This is used to generate a matrix that is of the correct
dimension.

ndcols

This function takes in the matrix of values of the MTR evaluated
over the grid generated for the audit procedure. The grid is
ordered according to the covariates first, and then by the
unobservables (this is done in by <code><a rd-options="" href="/link/genmonoA?package=ivmte&version=1.0.1" data-mini-rdoc="ivmte::genmonoA">genmonoA</a></code>). This
function takes the first difference of the unobservables within
each set of values for the covariates. This is sufficient to
generate the monotonicity constraint matrix.

The marginal treatment effect was introduced by Heckman and
Vytlacil (2005) <doi:10.1111/j.1468-0262.2005.00594.x> to provide a
choice-theoretic interpretation to instrumental variables models that
maintain the monotonicity condition of Imbens and Angrist (1994)
<doi:10.2307/2951620>. This interpretation can be used to extrapolate from
the compliers to estimate treatment effects for other subpopulations. This
package provides a flexible set of methods for conducting this
extrapolation. It allows for parametric or nonparametric sieve estimation,
and allows the user to maintain shape restrictions such as monotonicity. The
package operates in the general framework developed by Mogstad, Santos and
Torgovitsky (2018) <doi:10.3982/ECTA15463>, and accommodates either point
identification or partial identification (bounds). In the partially
identified case, bounds are computed using linear programming. Support for
three linear programming solvers is provided. Gurobi and the Gurobi R API
can be obtained from <http://www.gurobi.com/index>. CPLEX can be obtained
from <https://www.ibm.com/analytics/cplex-optimizer>. CPLEX R APIs 'Rcplex'
and 'cplexAPI' are available from CRAN. The lp_solve library is freely
available from <http://lpsolve.sourceforge.net/5.5/>, and is included when
installing either of its R APIs, 'lpSolve' or 'lpSolveAPI', which are
available from CRAN.

Joshua Shea

ivmte

Instrumental Variables: Extrapolation by Marginal Treatment
Effects

Alexander Torgovitsky

diffA function

This function takes in the matrix of values of the MTR evaluated
over the grid generated for the audit procedure. The grid is
ordered according to the covariates first, and then by the
unobservables (this is done in by <code><a rd-options='' href='genmonoA'>genmonoA</a></code>). This
function takes the first difference of the unobservables within
each set of values for the covariates. This is sufficient to
generate the monotonicity constraint matrix.

diffA: Taking first differences of constraint matrices

Description

Usage

Arguments

Value