Optimal Nonparametric Testing of Missing Completely at Random

---

Documentation for package ‘MCARtest’ version 1.1

• DESCRIPTION file.

Help Pages

Amatrix	Generate the matrix A, whose columns are the vertices of the marginal polytope.
AmatrixSparse	Generate the matrix A, whose columns are the vertices of the marginal polytope, as a sparse matrix.
Cimproved	Calculate the critical value for our improved test
col_index	A function indexing the columns of A
ConsMinkSumHrep	Calculate the H-representation of the consistent Minkowski sum
Csimple	Calculate the critical value for our simple test
EMiteration	Perform one step of the EM algorithm for finding the MLE under MCAR in a contingency table.
EquivalenceClass	Simplifies H-representation by exploiting symmetry
FuchsTest	Carry out Fuchs's test of MCAR in a contingency table, given complete and incomplete observations.
InconsMinkSumHrep	Calculate the H-representation of the general (possibly inconsistent) Minkowski sum
loglik0	Compute the log likelihood of a probability mass function, under MCAR, given complete and incomplete data
loglik1	Compute the log likelihood of a probability mass function, without assuming MCAR, given complete and incomplete data
MargPolyHrep	Calculate the H-representation of the marginal polytope
MLE	Compute the MLE under MCAR in a contingency table using the EM algorithm, given complete and incomplete observations.
ProjectionTest	Carry out a test of MCAR in a contingency table, given incomplete observations.
Rindex	A function computing the incompatibility index
RindexDual	A function computing the incompatibility index and associated closest joint mass function using the dual formulation
RoundErrors	Round errors in halfspace representations
row_index	A function indexing the rows of A