In dajmcdon/ubc-stat406-labs: Tutorials and labs for UBC Stat 406 in the 2020-2021 online year

See the decision_boundary() as a documented function.

library(knitr)
opts_chunk$set(echo=TRUE, fig.align='center', fig.width=4, fig.height=3,
               cache=TRUE, autodep=TRUE, cache.comments=FALSE,
               message=FALSE, warning=FALSE)

Analysis with the famed "iris" data

This is a very old, and by now canonical data set. It's in base R. Load it with data(iris).

1. Using this data, perform lda, logistic regression and qdausing only the first 2 variables. Note: logit for more than two classes is easiest with the multinom function in the nnet package. The other methods are in MASS.

data("iris")

1. Use the following function to view the discriminant boundaries for each of the methods above.

decisionplot <- function(model, predict_type = "class",
  resolution = 100, showgrid = TRUE, ...) {
  cl = iris[,5]
  k <- length(unique(cl))
  plot(iris[,1:2], col = as.integer(cl)+1L, pch = as.integer(cl)+1L, ...)
  # make grid
  r <- sapply(iris[,1:2], range, na.rm = TRUE)
  xs <- seq(r[1,1], r[2,1], length.out = resolution)
  ys <- seq(r[1,2], r[2,2], length.out = resolution)
  g <- cbind(rep(xs, each=resolution), rep(ys, time = resolution))
  colnames(g) <- colnames(r)
  g <- as.data.frame(g)
  ### guess how to get class labels from predict
  ### (unfortunately not very consistent between models)
  p <- predict(model, g, type = predict_type)
  if(is.list(p)) p <- p$class
  p <- as.factor(p)

  if(showgrid) points(g[[1]],g[[2]], col = as.integer(p)+1L, pch = ".")

  z <- matrix(as.integer(p), nrow = resolution, byrow = TRUE)
  contour(xs, ys, z, add = TRUE, drawlabels = FALSE,
    lwd = 2, levels = (1:(k-1))+.5)
  invisible(z)
}

1. Redo the analysis using all four covariates. Report the classification error of each method with two and four covariates.

dajmcdon/ubc-stat406-labs documentation built on Aug. 18, 2020, 1:23 p.m.