Fast Negative Binomial Regression, Estimate-Only (R Wrapper)
Source:R/helper_glm_fit.R
fast_negbin_regression.RdThis function provides a fast implementation of mean/dispersion-parameterized
negative binomial regression, wrapping a C++ backend
(fast_neg_bin_cpp; see that page, and
fast_dnbinom_mu_vec_cpp, for the full model). It returns point
estimates only (no standard errors) — see fast_negbin_regression_with_var
for the variance-computing counterpart. Columns 1 and 2 of X (conventionally
the intercept and treatment indicator) are always kept; this wrapper adds automatic,
silent handling of rank-deficient or numerically unstable covariate sets
beyond those first two columns: (1) if warm_start_params is supplied, a single
fit is attempted on the full matrix using the warm start, and its result is returned
if successful; (2) otherwise, upfront, any covariate columns (3 onward) found
rank-deficient by qr are dropped before the first fit attempt;
(3) if the C++ fit still fails (e.g. an L-BFGS line-search failure), covariates are
dropped one at a time, in reverse QR-pivot order (most redundant first), retrying
after each drop, until the fit succeeds or only the intercept and treatment columns
remain — at which point, if it still fails, this function stop()s with an
explicit error rather than returning a corrupted fit.
Usage
fast_negbin_regression(
X,
y,
optimization_alg = "lbfgs",
warm_start_params = NULL,
warm_start_fisher_info = NULL
)Arguments
- X
A numeric matrix of predictor variables. It is assumed that an intercept column (e.g., a column of ones) is already included in
Xif desired.- y
A numeric vector of the response variable, representing count data.
- optimization_alg
Optimization algorithm:
"lbfgs"(default) or"newton_raphson".- warm_start_params
Optional starting values for coefficients and
log_theta, passed straight through to the C++ backend for a single warm-started fit attempt (bypassing the QR-dropping retry sequence).- warm_start_fisher_info
Optional initial Fisher information matrix, used only together with
warm_start_params.
Value
A list containing the following components:
- b
A numeric vector of the estimated negative binomial regression coefficients \(\hat\beta\) (and
log_theta), from whichever (possibly covariate-reduced) fit in the retry sequence ultimately succeeded.- fisher_information
The C++ backend's returned Fisher information matrix for the successful fit.
See also
fast_negbin_regression_with_var for the variance-computing,
similarly retry-hardened wrapper; fast_neg_bin_cpp for the underlying
backend and full model documentation.
Examples
X = matrix(rnorm(100), 10, 10)
y = rpois(10, 2)
fast_negbin_regression(X, y)
#> $b
#> [1] -1.0493003 -0.2414721 0.8880010 1.2720706 1.4348667 0.7226149
#> [7] 0.1949020 -0.6832334 0.2999400 -0.2060604
#>
#> $fisher_information
#> [,1] [,2] [,3] [,4] [,5]
#> [1,] 1.598314e+01 -1.279838e+01 1.114382e+01 4.102916e+00 1.5153758199
#> [2,] -1.279838e+01 2.422699e+01 -7.808594e+00 -1.308196e-01 -7.5661633056
#> [3,] 1.114382e+01 -7.808594e+00 2.720672e+01 2.832963e+00 -2.5396020072
#> [4,] 4.102916e+00 -1.308196e-01 2.832963e+00 6.552712e+00 -0.5337676945
#> [5,] 1.515376e+00 -7.566163e+00 -2.539602e+00 -5.337677e-01 17.9996111621
#> [6,] -1.602372e+01 1.734974e+01 -9.850515e+00 -2.963110e-01 1.6955593369
#> [7,] -2.627289e+00 2.166753e+00 1.134921e+00 -6.399363e-01 -7.5759608398
#> [8,] -7.302258e+00 2.720078e+00 -3.044497e+00 3.414965e-01 18.4233360615
#> [9,] 1.487046e+00 -3.896146e+00 -1.943189e+00 -9.814531e-01 -8.2635348763
#> [10,] 1.286278e+01 -1.263959e+01 9.162636e+00 5.588362e-01 4.9106844081
#> [11,] -1.893954e-05 -2.167054e-05 -2.748439e-06 -2.353656e-06 0.0000137036
#> [,6] [,7] [,8] [,9] [,10]
#> [1,] -1.602372e+01 -2.627289e+00 -7.302258e+00 1.487046e+00 1.286278e+01
#> [2,] 1.734974e+01 2.166753e+00 2.720078e+00 -3.896146e+00 -1.263959e+01
#> [3,] -9.850515e+00 1.134921e+00 -3.044497e+00 -1.943189e+00 9.162636e+00
#> [4,] -2.963110e-01 -6.399363e-01 3.414965e-01 -9.814531e-01 5.588362e-01
#> [5,] 1.695559e+00 -7.575961e+00 1.842334e+01 -8.263535e+00 4.910684e+00
#> [6,] 2.298888e+01 -7.431071e-01 1.572883e+01 -5.251816e+00 -1.689993e+01
#> [7,] -7.431071e-01 1.207085e+01 -4.759070e+00 1.156822e+01 -1.574582e+00
#> [8,] 1.572883e+01 -4.759070e+00 3.111394e+01 -1.040008e+01 -4.323446e+00
#> [9,] -5.251816e+00 1.156822e+01 -1.040008e+01 1.982008e+01 -2.130090e+00
#> [10,] -1.689993e+01 -1.574582e+00 -4.323446e+00 -2.130090e+00 1.828977e+01
#> [11,] 1.177624e-05 -1.267319e-05 7.630130e-06 -1.171774e-05 -1.115984e-05
#> [,11]
#> [1,] -1.893954e-05
#> [2,] -2.167054e-05
#> [3,] -2.748439e-06
#> [4,] -2.353656e-06
#> [5,] 1.370360e-05
#> [6,] 1.177624e-05
#> [7,] -1.267319e-05
#> [8,] 7.630130e-06
#> [9,] -1.171774e-05
#> [10,] -1.115984e-05
#> [11,] 4.796574e-04
#>