EN
The paper presents the scope, methods and some results of the analysis of cyclical variability of morę than 100 economic variables pertinent to the cyclical movement of Poland’s economy over the period 1975-1998 (filled with monthly or ąuarterly data). The original database includes 107 individual variables and 17 composite reference indicators, all expressed as uniform volume indices (1992 avg. = 100). The research was accomplished in three phases: (a) decomposition of time series using X11-ARIMA, coupled with seasonal adjustment and MCD m.a. smoothing; (b) analysis of cyclical components isolated by PAT procedure (or altematively, by linear regression); (c) analysis of cross-correlation against the reference cycle and the determination of leads and lags. The number of variables was successively reduced, according to the adopted selection criteria. The finał set includes ca. 60 variables accepted for the construction of composite indicators of business activity. A subsidiary database including 60 variables from business surveys has been established in the last two years to be used in composite qualitative indicators reflecting the business climate.