From the plot of the time series data mentioned in the preceding session, we can clearly see the seasonality patterns. Now we will see the ways to extract the seasonality component and trend component. This can be implemented using the stl function also called Seasonal Decomposition of Time series by Loess. This method decomposes the dataset into seasonal, trend, and irregular components using the Loess method. Loess is a method of estimating nonlinear relationships.
First, we will consider a subset of data from the time series dataset for a better visualization of the components. We can extract a subset of data using the window function, which takes the time series data itself as an input along with the start and end dates. In the start=c(2007,1) parameter, 2007 is the start year and 1 is the start month in the year 2007. Hence, the following code creates a subset ranging from 2007 to 2009. After subsetting the dataset, we will use the plot function to visualize the output...