Forecasting the state of the business cycle, including that in the transport sector, is a difficult task. In the case of transport, this is primarily due to its strong coupling with other sectors of the economy.
This is not the only reason – the long-term, “ordinary” relationship between the transport sector and its economic environment is additionally overlaid by factors whose impact can be both evolutionary and more abrupt, leading to structural changes. Examples include, on the one hand, rising energy prices, and on the other, changes in the volume and structure of transport accompanying economic development that favors knowledge and new technologies. It is also worth mentioning the possible further liberalization of the transport market (primarily rail), the development of transport infrastructure, the drive to integrate different modes of transport into logistics chains, the widespread use of intelligent transport systems, the need to take environmental standards into account, etc. These and other factors can significantly change the structure of demand (e.g., by reducing demand for raw materials and increasing demand for services) and, consequently, cause the relationships between transport and its environment to be non-stationary. Even just because of the issues mentioned above, constructing sufficiently good forecasts of transport market development is difficult and requires taking into account potentially probable phenomena, events, and processes that often have no precedent in the history of road transport in Poland and the world.
The purpose of the considerations contained in this paper is to determine short-term forecasts of road transport sector characteristics: transport volume and transport performance. Quarterly data have been published by the Central Statistical Office (GUS) only recently, so the relevant time series are not very long (covering the period from the first quarter of 2004). For this reason, emphasis has been placed on “soft” data, namely business climate indicators in transport, which have been determined since the beginning of 1997 at the Motor Transport Institute.
Due to the fact that the forecast horizon is relatively short – covering the next 8 quarters (the years 2014 and 2015) – the aforementioned global structural changes were not taken into account. The tools used to construct the forecasts are selected time series models. Thus, the determined forecasts will, by necessity, be based only on the observation of events, factors, and relationships shaping changes in the transport market to date.
Forecasts for each of the mentioned variables are determined based on a number of time series models and three procedures used to construct combined forecasts. We describe these issues in the next two parts of the article. The last part is devoted to the presentation of the results. The relevant calculations were performed using procedures written in the R environment.
The article contains 22,660 characters.
Source: Logistyka Magazine