Constructing a Composite Leading Indicator for the Czech Business Cycle: A Novel Multi-Criteria TOPSIS Approach
Аннотация
A newly constructed composite leading indicator for the Czech business cycle is introduced, while building upon and enhancing the OECD’s methodological framework. Central to the contribution is the formalisation of the selection process for candidate component series. An existing point-based evaluation method is compared with an alternative approach that utilises the multi-criteria decision analysis technique known as TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution), which allows for the simultaneous assessment of multiple criteria. Results indicate that the TOPSIS method surpasses simple point assignment in both timeliness and stability of the final indicator. Additionally, various weighting schemes are applied to evaluate different performance metrics, indicating that the use of median lead times, rather than mean lead times, yields superior results. The study further investigates the optimal number of component series included. A secondary application of the TOPSIS method across indicators with differing compositions is recommended, selecting the best performer as optimal. Finally, the indicator is tested under simulated real-time conditions, demonstrating its capacity to anticipate turning points in the Czech GDP with an average lead of five months.