Calculate and plot a rolling average
The final baseball indicator you'd like to generate is the L10, or the moving win/loss average from the previous ten games. While the cumulative win/loss average tells you how the team is doing overall, the L10 indicator provides a more specific picture of the team's recent performance. Beyond the world of sports, this measure is comparable to a financial indicator focused on recent portfolio performance.
To generate a rolling win/loss average, return to the rollapply()
command used in the previous chapter. In this case, you'll want to apply the mean
function to the last 10
games played by the Red Sox at any given time during the 2013 season.
The redsox_xts
object, including the win_loss
column, is available in your workspace.
This exercise is part of the course
Case Study: Analyzing City Time Series Data in R
Exercise instructions
- Generate a new xts object containing only the 2013 season. Call this object
redsox_2013
. - Use
rollapply()
to calculate yourlastten_2013
indicator based on thewin_loss
column inredsox_2013
. Set thewidth
equal to10
to include the last ten games played by the Red Sox and set theFUN
argument tomean
to generate an average of thewin_loss
column. - Use
plot.xts()
to view your new indicator during the 2013 season. Leave theylim
argument as is in the prewritten code.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Select only the 2013 season
redsox_2013 <- ___["___"]
# Use rollapply to generate the last ten average
lastten_2013 <- rollapply(___$___, width = ___, FUN = ___)
# Plot the last ten average during the 2013 season
plot.xts(___, ylim = c(0, 1))