Rolling functions

Case Study: Analyzing City Time Series Data in R

Lore Dirick

Manager of Data Science Curriculum at Flatiron School

Discrete windows

Split the data according to period

unemployment_yrs <- split(unemployment, f = "years")

Apply function within period

unemployment_yrs <- lapply(unemployment_yrs, cummax)

Bind new data into xts object

unemployment_ytd <- do.call(rbind, unemployment_yrs)

rollapply() applies a function to a rolling window

unemployment_avg <- rollapply(unemployment,
                              width = 12,
                              FUN = mean)

plot showing rollapply() used to get 12-month window means

Case Study: Analyzing City Time Series Data in R