In this blog post graphic methods are used to describe tourism time series in Serbia.
Two types of a line graph could be used to show changes in tourist demand over time. The first displays all three time series (total, domestic and foreign tourists) on a single graph. This graph shows the relative position of time series to each other. However, if the time series levels are disproportional in size then on this graph is difficult to spot patterns in time series.
The second graph displays each series on a separate panel. On such graph any persistent pattern related to trend or seasonal factors should be clearly visible.
Figure 1 shows line graphs of tourists arrival and overnight stay in Serbia.
The first graph in Figure 1 shows a steady increasing trend in the foreign tourists arrival, while the domestic tourists arrival time series shows decreasing trend until 2014. Since then both time series made almost equal contribution to the total tourist arrival series resulting in a rapidly increasing trend in this time series. This increasing trend in the foreign tourist arrival series had also an impact on the seasonal pattern of the total tourists arrival. This could be seen if we look how the peaks in total tourists arrival series are changing, particularly after 2014. The situation with the overnight stay series is a bit different, because the domestic tourists stay overnight on average about two times longer than foreign tourists.
Both total and domestic tourists overnight stay series show similar but not quite stable seasonal patterns. This should be more clearly visible on the separate graphs in Figure 2.
On the graphs in Figure 2 the seasonal patterns of domestic and foreign tourists arrival and overnight stay are very similar. Trends, as another feature of domestic and foreign tourists arrival and overnight stay series are quite different. While domestic tourists time series display a decreasing trend until the end of 2014, then an increase after that, foreign tourists time series show an increasing trend in the whole period. These differences between time series will be further examine after decomposing time series into trend, season and residual series.