Polyclass: polychotomous regression and multiple classification

This function summarizes both the stepwise selection process of the model fitting by polyclass, as well as the final model that was selected

# S3 method for class 'polyclass'
summary(object, ...) 
# S3 method for class 'polyclass'
print(x, ...)

Arguments

object,x

polyclass object, typically the result of polyclass.

...

other arguments are ignored.

Value

These function summarize a polyclass fit identically. They also give information about fits that could have been obtained with other model selection options in polyclass.

References

Charles Kooperberg, Smarajit Bose, and Charles J. Stone (1997). Polychotomous regression. Journal of the American Statistical Association, 92, 117–127.

Charles J. Stone, Mark Hansen, Charles Kooperberg, and Young K. Truong. The use of polynomial splines and their tensor products in extended linear modeling (with discussion) (1997). Annals of Statistics, 25, 1371–1470.

Author

Charles Kooperberg clk@fredhutch.org.

See also

polyclass, plot.polyclass, beta.polyclass, cpolyclass, ppolyclass, rpolyclass.

Examples

data(iris)
fit.iris <- polyclass(iris[,5], iris[,1:4])
#> warning - model size was reduced
summary(fit.iris)
#> ========================POLYCLASS summary=======================
#> The fit was obtained with
#>   polyclass(data = iris[, 5], cov = iris[, 1:4])
#> There were 3 classes and 4 covariates.
#> There were 150 cases.
#> 
#> The model selection was carried out using AIC.
#> The penalty was the default, log(150 )=5.01 .
#> The model had dimension 6 , log-likelihood -10.28 and AIC -50.63 .
#> 
#> The locations of the knots:
#>   Number
#> 1      0
#> 2      0
#> 3      0
#> 4      0
#> 
#>   There are 3 basis functions, summarized below:
#>   dim1 knot1 dim2 knot2 Class 1 Class 2 Class 3
#> 1   NA    NA   NA    NA 150.293  45.263       0
#> 2    3    NA   NA    NA -48.334  -5.752       0
#> 3    4    NA   NA    NA -11.723 -10.449       0
#> The first basis function is the constant function. For all others,
#> the first column and the third column indicate on which covariates
#> that basis function depends. If the third column is NA, the basis
#> function depends on only one covariate.
#> For the nonconstant basis functions the second and the fourth column
#> indicate on which knot the function depend. If these columns are NA,
#> the basis function is linear in this covariate.
#> The remaining columns give the coefficients.
#> 
#> ================================================================
#> The influence of the penalty parameter is summarized below:
#>  dim     AIC l-lik-trn loss-trn sq-err-trn A/D pen-min pen-max
#>    2 339.605    -1.099    0.667      0.444   1 148.081     Inf
#>    4  53.464    -0.111    0.040      0.033   0   6.429 148.081
#>    6  50.627    -0.069    0.040      0.023   0   3.649   6.429
#>    8  53.351    -0.044    0.020      0.015   0   0.000   3.649
#>   10  63.372    -0.044    0.020      0.015   1   0.000   0.000
#> ================================================================
#> The importance-anova decomposition is:
#>  Cov-1 Cov-2 Percentage
#>     NA    NA      29.13
#>      3    NA      70.10
#>      4    NA       0.78
#> ================================================================