classify_multiclass {sim2Dpredictr}

Classify subjects based on predicted probabilities for each class

Description

Classify subjects based on predicted probabilities for each class. The predicted probabilities can be input by the user or calculated within function using parameter estimates and design matrix for a multinomial regression model.

Usage

classify_multiclass(
  predicted.probs = NULL,
  category.names,
  keep.probs = TRUE,
  B = NULL,
  X = NULL,
  X.incl.X0 = FALSE
)

Arguments

predicted.probs

A matrix where the number of rows is equal to the number of subjects and the number of columns equals the number of categories. predicted.probs[i, j] contains the probability that subject i belongs to category j.

category.names

A vector containing the names of each category. The order of names should match the order of columns in predicted.probs; correspondingly, the length of the vector should equal the number of columns in predicted.probs.

keep.probs

Logical. When TRUE, the output is data frame consisting of the information in predicted.probs with an additional column predicted.class that contains the predicted class for each subject. When FALSE, a vector of the predicted classes is returned.

B

A list, each element of which contains a parameter vector. The list should have length V - 1, i.e., should contain parameter values associated with all categories except the reference category, following Agresti (2007). Alternatively, B may be a list of length V if one desires to specify parameters for every category, i.e., the over-parameterized model used in Friedman (2010).

X

A matrix, each row of which contains subject covariate/predictor values.

X.incl.X0

Logical. When TRUE, X should contain column of 1's for the intercept. Otherwise, a column of 1's is generated internally. Default is FALSE.

Details

Classification for each subject is determined based on the category with highest predicted probability.

Value

Depending on the option selected for keep.probs, returns a data frame or vector.

Examples

## number of categories
vt <- 3

## covariate values
xt <- matrix(rnorm(10 * 2), ncol = 2, nrow = 10)

## list of parameter vectors (over-parameterized model)
bu <- list(b1 = c(0, 0.25, 0.25),
           b2 = c(0, -0.25, -0.25),
           b3 = c(0, 0.25, -0.25))

## subject specific probabilities for each category 
## (over-parameterized model)
prp <- generate_multinom_probs(V = vt, X = xt, B = bu)

classify_multiclass(predicted.probs = prp, 
                    category.names = c("A", "B", "C"))

## generate predicted probabilities within function
classify_multiclass(predicted.probs = NULL, 
                    category.names = c("A", "B", "C"),
                    X = xt, B = bu)