Extract coefficients from NMF models

Returns the coefficients data frame from a fitted NMF model that has been passed through an inference function (nmfkc.inference, nmf.rrr.inference, nmfre.inference).

If inference has not been run, returns the parameter matrix \(C\) (\(\Theta\)) directly.

For nmf.ffb objects, returns \(C_2\) (exogenous block) as fallback.

Usage

# S3 method for class 'nmf'
coef(object, ...)

# S3 method for class 'nmf.ffb'
coef(object, ...)

Arguments

object

A fitted model object of class "nmf", "nmfkc", "nmfae", "nmfre", or "nmf.ffb".

...

Not used.

Value

A data frame of coefficients (if inference was performed), or the parameter matrix \(C\).

See also

nmfkc.inference, nmf.rrr.inference, nmfre.inference, nmf.ffb.inference

Examples

Y <- matrix(cars$dist, nrow = 1)
A <- rbind(1, cars$speed)
result <- nmfkc(Y, A, rank = 1)
#> Y(1,50)~X(1,1)C(1,2)A(2,50)=XB(1,50)...
#> 0sec
coef(result)  # returns C matrix
#>             Cov1    Cov2
#> Basis1 0.1563808 2.89964

result2 <- nmfkc.inference(result, Y, A)
coef(result2)  # returns coefficients data frame
#>    Basis Covariate  Estimate        SE       BSE    z_value      p_value
#> 1 Basis1      Cov1 0.1563808 6.5123713 3.3507470 0.02401288 4.904212e-01
#> 2 Basis1      Cov2 2.8996401 0.4642225 0.4391579 6.24622937 2.102398e-10
#>     CI_low   CI_high on.boundary
#> 1 0.000000 10.740750       FALSE
#> 2 2.129304  3.942492       FALSE