Skip to contents

Statistical inference for an NMF-GMM fit (given basis)

Source: R/nmf.gmm.R
nmf.gmm.inference.Rd

This function is experimental and still under development. The interface may change in future versions: argument names, defaults and the contents of the returned object are not yet stable. Code written against it today may need adjusting after an update. The rest of the package does not carry this caveat.

nmf.gmm.inference adds Wald inference on the covariate-coefficient matrix \(C\) (\(=\Theta\)) of a fitted nmf.gmm object, conditional on the estimated basis \(\hat X\) and mixture. It reports the analytic outer-product (mixture-information) standard error — the coverage-calibrated choice at \(K>1\) — together with a wild-bootstrap standard error and percentile confidence interval, mirroring nmfre.inference / nmfkc.inference. Supported for the tied-covariance variant.

Usage

nmf.gmm.inference(object, Y, A = object$A, ...)

Arguments

object

A fitted "nmf.gmm" object (with cov = "tied").

Y

The data matrix used in the fit.

A

The covariate matrix used in the fit. Default object$A.

...

Additional arguments, named as in the other wild-bootstrap inference functions of the package: wild.B (bootstrap replicates, default 500), wild.level (confidence level, default 0.95) and seed (default 123).

Value

object with class c("nmf.gmm.inference", "nmf.gmm") and added components: coefficients (data frame with Basis, Covariate, Estimate, SE, BSE, z_value, p_value, CI_low, CI_high), C.se.outer, C.se.sandwich, C.se.boot (Q x R matrices) and Xi (\(= X C\)).

See also

nmf.gmm, nmf.gmm.select

Examples

# \donttest{
set.seed(1)
Y <- matrix(abs(rnorm(20 * 60)) + 1, 20, 60); A <- rbind(1, rnorm(60))
fit <- nmf.gmm(Y, A, rank = 2, K = 2)
fit <- nmf.gmm.inference(fit, Y, A, wild.B = 300)
head(fit$coefficients)
#>    Basis Covariate    Estimate        SE       BSE     z_value   p_value
#> 1 Basis1      Cov1 36.27851926 0.7824778 0.3225164 46.36364255 0.0000000
#> 2 Basis2      Cov1  0.14127734 0.5034434 0.3956258  0.28062206 0.7790003
#> 3 Basis1      Cov2 -0.06061948 0.7665946 0.2878565 -0.07907633 0.9369719
#> 4 Basis2      Cov2  0.27859573 0.5495305 0.2928593  0.50697046 0.6121755
#>       CI_low    CI_high
#> 1 35.6346689 36.9039131
#> 2 -0.6222280  0.8555773
#> 3 -0.5826109  0.4764388
#> 4 -0.3218638  0.8484674
# }