Short-term economic indicators play an important role in the assessment of current cyclical situations and in the establishment of forecasts. Broadly, two types of short-term indicators can be distinguished: qualitative indicators, reflecting businessmen's subjective assessment of the cyclical situation (e.g., production prospects or judgements on orderbooks), and quantitative indicators, reflecting past developments of production or employment. The usefulness of qualitative indicators rests on their reliability to approximate the possible evolution of the quantitative ones.

The paper shows the results of a pilot study for six industrial sectors in seven OECD Member countries (Canada, Japan, Norway, Spain, Sweden, Switzerland and the United States) whereby short-term qualitative indicators are used to "nowcast" a quantitative indicator, the production index. The objective is to enhance the timeliness of short-term industrial statistics through estimation of data points for the most recent periods for which they are not yet available. The results of the pilot study indicate that: a) qualitative variables are clearly useful in forecasting indices of production; b) at the same time, they are not by themselves sufficient to provide good estimates of quantitative indices; c) the estimated time-series equations trace well the actual values of the production indices as comparisons of actual and fitted values for seven countries show; d) forecasting errors of the models are within reasonable bounds although there are clear differences across countries and sectors.