apress / advanced-r-statistical-programming-and-data-models

Source Code for 'Advanced R Statistical Programming and Data Models' by Matt Wiley and Joshua F. Wiley

Apress Source Code

This repository accompanies Advanced R Statistical Programming and Data Models by Matt Wiley and Joshua F. Wiley (Apress, 2019).

Download the files as a zip using the green button, or clone the repository to your machine using Git.

Releases

Release v1.0 corresponds to the code in the published book, without corrections or updates.

Contributions

See the file Contributing.md for more information on how you can contribute to this repository.

advanced-r-statistical-programming-and-data-models's Issues

data from "http://doi.org/10.3886/ICPSR04690.v7" is not available

It appeared that data from "http://doi.org/10.3886/ICPSR04690.v7" is not available any more. Could you please share the dataset you used here or some other website? Thanks!

Jeff

Ch. 7 ML: Unsupervised, page 267: plots list filled with last ggplot only

I was unable to reproduce the results shown on page 268 using the suggested code

     p1 <- ggplot(d, aes(NYGN, SPAD))
  wgss <- vector('numeric', 8)
  plots <- vector('list', 9)

for(i in 2:9) {

 km <- kmeans(x, centers = i)
    
  wgss[i-1] <- km$tot.withinss
  plots[[i-1]] <- p1 + geom_point(aes(color = factor(km$cluster))) +
                     scale_color_viridis(discrete = TRUE) +
                     theme(legend.position = 'none') +
                     ggtitle(paste('kmeans with', i, 'centers'))

}
 plots[[9]] <- ggplot() + geom_point(aes(x = 2:9, y = wgss)) + 
                  xlab('Number of Clusters') +
                  ylab('Within SS') +
                  ggtitle('Scree Plot')

do.call(plot_grid, c(plots, ncol = 3))

As result, only the 8-th plot fills the first 8 components of plots list, with the 9- th being the scree plot:

Reproduction of results was possible by modifying the code as follows:

       p1 <- ggplot(d, aes(NYGN, SPAD))
    wgss <- vector('numeric', 8)
    plots <- vector('list', 9)
clusters <- vector('list', 8)


for(i in 2:9) {

 km <- kmeans(x, centers = i)
  wgss[i-1] <- km$tot.withinss
  clusters[[i-1]] <- km$cluster
 
}
scree <- ggplot() + geom_point(aes(x = 2:9, y = wgss)) + 
                  xlab('Number of Clusters') +
                  ylab('Within SS') +
                  ggtitle('Scree Plot')

plots <- lapply(2:9, 
             function(i) {p1 + 
                         geom_point(aes(color = factor(clusters[[i-1]]))) +
                         scale_color_viridis(discrete = TRUE) +
                         theme(legend.position = 'none') +
                         ggtitle(paste('kmeans with', i, 'centers'))
                             }
                      )

plots <- c(plots, list(scree))

do.call(plot_grid, c(plots, ncol = 3))

which yielded:

UPDATE: I have missed mapping the geom_point with aes_ in my first code example. With aes_ I was able to reproduce the results shown on page.