Simen Gaure

Transforms away factors with many levels prior to doing an OLS. Useful for estimating linear models with multiple group fixed effects, and for estimating linear models which uses factors with many levels as pure control variables. Includes support for instrumental variables, conditional F statistics for weak instruments, robust and multi-way clustered standard errors, as well as limited mobility bias correction.

Contains methods for creating multivariate/multidimensional interpolations of functions on a hypercube. If available through fftw3, the DCT-II/FFT is used to compute coefficients for a Chebyshev interpolation. Other interpolation methods for arbitrary Cartesian grids are also provided, a piecewise multilinear, and the Floater-Hormann barycenter method. For scattered data polyharmonic splines with a linear term is provided. The time-critical parts are written in C for speed. All interpolants are parallelized if used to evaluate more than one point.

Estimation of piecewise constant mixed proportional hazard competing risk model with NPMLE. The model is described in S. Gaure et al. (2007) <doi:10.1016/j.jeconom.2007.01.015>, J. Heckman and B. Singer (1984) <doi:10.2307/1911491>, and B.G. Lindsay (1983) <doi:10.1214/aos/1176346059>.

R wrappers around the cubature C library of Steven G. Johnson for adaptive multivariate integration over hypercubes and the Cuba C library of Thomas Hahn for deterministic and Monte Carlo integration. Scalar and vector interfaces for cubature and Cuba routines are provided; the vector interfaces are highly recommended as demonstrated in the package vignette.