In this chapter, an attempt to analyze ąuality of predictions of industrial enterprises is described. Data originates from monthly questionnaires conducted by the Institute of Economic Developments Warsaw School of Economics. Usually, two groups of questions are included in questionnaires of overall economic situation. First group concerns current values of various economic variables, second one - changes in these values in near futurę. Expectations of respondents create a certain expert system of predictions. The obvious question is: what is the quality of these predictions? In this chapter, attention has been focused on industrial output which plays a major part in building a comprehensiye coefficient of economic conditions of industry.
In the paper we investigate possibility of using the Viterbi paths to analyze two-dimensional macroeconomic time series. We build a two-dimensional Gaussian Markov-switching model with a four-state hidden Markov chain. The model is tested with two pairs of monthly indexes of industrial production for: Poland vs. France, and Poland vs. Germany. The most likely sequence of states of the hidden Markov chain is found for each pair. We compare that sequence with analogous sequences determined for a one-dimensional model with a two-state hidden Markov chain. The results of the comparison suggests the four state Viterbi path provides more valuable information about business cycle synchronization between the two economies than two separate two-state Viterbi paths.
The aim of the paper is to show that turning points detection can be treated as a problem of pattern recognition. In the paper there are presented the results of applying normal hidden Markov models to a number of survey balances. Beyond a classical two-scale assessment of business activity a slightly more fuzzy classification of states is considered. To determine periods of unclear or difficult to evaluate situation unobservable Markov chains with three and four states are introduced. The outputs of the Viterbi algorithm, i.e. the most likely paths of unobservable states of Markov chains, are a basis of the proposed classification. The comparison of these paths with the business cycle turning points dated by OECD is described. The results obtained for three- and four-state Markov chains are close to those established in the references time series and seem to improve the speed with which, especially downshifts, are signaled. Furthermore, these results are more favorable than outcomes provided by conventional two-state models. The method proposed in this paper seems to be a very effective tool to analyze results of business tendency surveys, in particular, when multistate Markov chains are considered. Moreover, proposed decompositions allow an easy comparison of two time series as far as turning point are concerned. In the paper survey balances are compared with ‘hard’ economic data such as sold manufacturing production. The results confirm the accuracy of assessment provided by survey respondents.