When working with time-series data, there are a number of data type formats to choose from and use for conversion. We have already used two of these formats, of which there are three that are most widely used. Let's briefly review these data types before moving on to our deep learning model.
When we wanted to add actual data as an overlay to our ARIMA model plot, we used the ts function to create a time-series data object. For this object, the index values must be integers. In the case of using the autolayer function with the arima plot, a time-series data object is required. This is one of the more simple time-series data types and it will look like a vector in your Environment tab. However, this only works with regular time series.
Another data type is zoo. The zoo data type will work with regular and irregular time series, and...
