Time series analysis

Consider the following time series plot of daily morning gold prices, based on the gold data from the forecast package:

library(forecast) 

autoplot(gold) 

We’d like to add a note about the spike due to incorrect value to the plot. First, we wish to find out on which day the spike appears: 

spike_date <- which.max(gold) 

To add a circle around that point, we add a call to annotate to the plot: 

autoplot(gold) + 

     annotate(geom = “point”, x = spike_date, y = gold[spike_date], 

              size = 5, shape = 21, colour = “red”, fill = “transparent”) 

Using the figure created above and the documentation for annotate, do the following: 

1. Add the text “Incorrect value” next to the circle.
2. Create a second plot where the incorrect value has been removed.
3. Read the documentation for the geom geom_hline. Use it to add a red reference line to the plot, at y=400.