Add a continuous rolling average to unemployment data
In addition to discrete measures such as year-to-date sums, you may be interested in adding a rolling sum or average to your time series data.
To do so, let's return to your monthly unemployment data. While you may be interested in static levels of unemployment in any given month, a broader picture of the economic environment might call for rolling indicators over several months.
To do this, you'll use the rollapply() command, which takes a time series object, a window size width, and a FUN argument to apply to each rolling window.
This exercise is part of the course
Case Study: Analyzing City Time Series Data in R
Exercise instructions
- Use
rollapply()to calculate the rolling yearly average US unemployment. Be sure to specify theuscolumn of yourunemploymentdata, set thewidthargument to the appropriate number of monthly periods, and set theFUNargument tomean. Save your rolling average into yourunemploymentobject asyear_avg. - Plot your two indicators of US unemployment (
usandyear_avg) usingplot.zoo(). Set theplot.typeargument to"single"to place both measures in the same panel.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Use rollapply to calculate the rolling yearly average US unemployment
unemployment$year_avg <- rollapply(___$___, width = ___, FUN = ___)
# Plot all columns of US unemployment data
plot.zoo(unemployment[, c("___", "___")], plot.type = "___", lty = lty, lwd = lwd)