Update to StatsModels and StatsBase changes

Change instances of stderr to stderror for StatsBase v0.22.0

Replace explicit calls to a Formula constructor with calls to @eval(@formula(...)) to take into account changes in StatsModels v0.2.4

Deprecate lmm and glmm functions

• Deprecate lmm and glmm constructor functions in favor of explicit LinearMixedModel and GeneralizedLinearMixedModel constructors
• Add fit(LinearMixedModel, ...) and fit(GeneralizedLinearMixedModel, ...) methods
• Adjust docs for new conventions

Update to BlockArrays v0.3.0

v0.16.3

Update to BlockArrays v0.3.0, ambiguous stderr

Update benchmarks

Update benchmark code to PkgBenchmark v0.1.0

Fixes for #116 and #117

• Add a method for αβAc_mul_B! needed when evaluating ranef on a model with nested, vector-valued random effects (#116)

• Add a length method for MatrixTerm (#117)

Allow rank-deficient fixed-effects model matrices

• Extend the MatrixTerm type with a piv vector and rank integer scalar. Use a pivoted Cholesky factorization to determine the computational rank of X'X.

• fixef gains an additional, logical argument pivoted that defaults to true.

• coef always returns the fixed-effects coefficients in the original order. Coefficients for columns found to be linearly dependent are set to -0.0 so that X*coef(m) is the fitted value that would be obtained via projection.

• add the 1-argument methods for StatsBase.predict (the result is equivalent to StatsBase.fitted).

Bump StatsBase and StatsModels requirements

The release of StatsBase v0.19.2 resulted in this package failing to precompile because StatsBase and StatsModels both exported coefnames. StatsModels v0.2.0 resolved this problem. The main purpose of this release is to require these versions of StatsBase and StatsModels.

The dof and nobs methods have been updates and tests added.

Some of the documentation files have been updated by re-weaving the .jmd sources. Those that depend on the Gadfly package cannot be updated until a new version of Gadfly is released.

Require julia v0.6.1

In 0.6.0 the colblocks member of the BlockedSparse type could not be created because a reshaped contiguous view of a vector was not recognized as a StridedMatrix.

Use DataFrames 0.11, CategoricalArrays, StatsModels

Update the code to use v0.11.1 or later of DataFrames, CategoricalArrays, and StatsModels packages.

Use BlockedSparse type to reduce allocation

• Reduce memory allocation and garbage collection in models with multiple vector-valued random effects
• Add tests and test coverage
• Add benchmarks