This function calculates the weight for each observation in the data matrix x in order to calculate the covariance matrices employed in the HDRDA classifier, implemented in rda_high_dim().

rda_weights(x, y, lambda = 1)

Arguments

x

Matrix or data frame containing the training data. The rows are the sample observations, and the columns are the features. Only complete data are retained.

y

vector of class labels for each training observation

lambda

the RDA pooling parameter. Must be between 0 and 1, inclusively.

Value

list containing the observations for each class given in y

References

Ramey, J. A., Stein, C. K., and Young, D. M. (2013), "High-Dimensional Regularized Discriminant Analysis."