Calculate Variance-Covariance Matrix for mixpoissonreg Objects

Returns the variance-covariance matrix of the parameters for fitted mixed Poisson regression models. The parameters argument indicates for which parameters the variance-covariance matrix should be computed, namely, 'mean' for mean-relatex parameters or 'precision' for precision-related parameters.

# S3 method for mixpoissonreg
vcov(object, parameters = c("all", "mean", "precision"), ...)

Arguments

object	an object of class "mixpoissonreg" containing results from the fitted model.
parameters	a string to determine which coefficients should be extracted: 'all' extracts all coefficients, 'mean' extracts the coefficients of the mean parameters and 'precision' extracts coefficients of the precision parameters.
...	further arguments passed to or from other methods.

Value

A matrix containing the covariance matrix of a mixpoissonreg object.

Examples

# \donttest{
data("Attendance", package = "mixpoissonreg")

daysabs_fit <- mixpoissonreg(daysabs ~ gender + math +
prog | gender + math + prog, data = Attendance)
vcov(daysabs_fit)
#>                            (Intercept)    gendermale          math
#> (Intercept)               2.173024e-02 -5.918840e-03 -2.218622e-04
#> gendermale               -5.918840e-03  1.391630e-02  1.824148e-05
#> math                     -2.218622e-04  1.824148e-05  5.369261e-06
#> progAcademic             -1.051474e-02 -1.025833e-03  6.643646e-06
#> progVocational           -5.911765e-03 -1.865578e-03 -1.019043e-04
#> (Intercept).precision     3.988849e-04 -4.684298e-04  8.386718e-06
#> gendermale.precision      7.371292e-04 -5.312241e-04 -3.678565e-05
#> math.precision            1.332653e-05 -3.548041e-05 -1.117223e-07
#> progAcademic.precision   -7.582598e-04  2.037610e-03  8.706788e-07
#> progVocational.precision -3.241746e-03  4.181426e-03  4.626451e-05
#>                           progAcademic progVocational (Intercept).precision
#> (Intercept)              -1.051474e-02  -5.911765e-03          3.988849e-04
#> gendermale               -1.025833e-03  -1.865578e-03         -4.684298e-04
#> math                      6.643646e-06  -1.019043e-04          8.386718e-06
#> progAcademic              1.746197e-02   1.062915e-02          2.197079e-04
#> progVocational            1.062915e-02   3.043066e-02         -3.347031e-03
#> (Intercept).precision     2.197079e-04  -3.347031e-03          1.178159e-01
#> gendermale.precision      7.818101e-04   3.612138e-03         -1.737732e-02
#> math.precision           -8.312163e-06   5.423776e-05         -7.945786e-04
#> progAcademic.precision   -2.333842e-04  -2.563226e-04         -7.751125e-02
#> progVocational.precision -2.082009e-04  -2.350747e-03         -6.340653e-02
#>                          gendermale.precision math.precision
#> (Intercept)                      7.371292e-04   1.332653e-05
#> gendermale                      -5.312241e-04  -3.548041e-05
#> math                            -3.678565e-05  -1.117223e-07
#> progAcademic                     7.818101e-04  -8.312163e-06
#> progVocational                   3.612138e-03   5.423776e-05
#> (Intercept).precision           -1.737732e-02  -7.945786e-04
#> gendermale.precision             4.149055e-02   4.782230e-05
#> math.precision                   4.782230e-05   1.748048e-05
#> progAcademic.precision          -3.021667e-03   5.731643e-05
#> progVocational.precision        -6.910452e-03  -2.345692e-04
#>                          progAcademic.precision progVocational.precision
#> (Intercept)                       -7.582598e-04            -3.241746e-03
#> gendermale                         2.037610e-03             4.181426e-03
#> math                               8.706788e-07             4.626451e-05
#> progAcademic                      -2.333842e-04            -2.082009e-04
#> progVocational                    -2.563226e-04            -2.350747e-03
#> (Intercept).precision             -7.751125e-02            -6.340653e-02
#> gendermale.precision              -3.021667e-03            -6.910452e-03
#> math.precision                     5.731643e-05            -2.345692e-04
#> progAcademic.precision             9.331766e-02             7.597928e-02
#> progVocational.precision           7.597928e-02             1.182064e-01
vcov(daysabs_fit, parameters = "mean")
#>                  (Intercept)    gendermale          math  progAcademic
#> (Intercept)     0.0217302395 -5.918840e-03 -2.218622e-04 -1.051474e-02
#> gendermale     -0.0059188402  1.391630e-02  1.824148e-05 -1.025833e-03
#> math           -0.0002218622  1.824148e-05  5.369261e-06  6.643646e-06
#> progAcademic   -0.0105147423 -1.025833e-03  6.643646e-06  1.746197e-02
#> progVocational -0.0059117646 -1.865578e-03 -1.019043e-04  1.062915e-02
#>                progVocational
#> (Intercept)     -0.0059117646
#> gendermale      -0.0018655785
#> math            -0.0001019043
#> progAcademic     0.0106291493
#> progVocational   0.0304306592
# }

daysabs_prog <- mixpoissonreg(daysabs ~ prog, data = Attendance)
vcov(daysabs_prog)
#>                         (Intercept)  progAcademic progVocational
#> (Intercept)            2.720872e-02 -2.720872e-02  -2.720872e-02
#> progAcademic          -2.720872e-02  3.402708e-02   2.720872e-02
#> progVocational        -2.720872e-02  2.720872e-02   3.999917e-02
#> (Intercept).precision  1.531513e-14  6.568027e-15  -3.236057e-15
#>                       (Intercept).precision
#> (Intercept)                    1.531513e-14
#> progAcademic                   6.568027e-15
#> progVocational                -3.236057e-15
#> (Intercept).precision          1.045938e-02

Calculate Variance-Covariance Matrix for `mixpoissonreg` Objects

Arguments

Value

See also

Examples