Escobal, J. y Torres, J. (2002). Un sistema de indicadores líderes del nivel de actividad para la economía peruana. Lima: GRADE. Documento de trabajo, 39.

The purpose of this paper is to present a methodology to construct a Composite Leading Indicator of Gross Domestic Product (GDP). The indicator aggregates a set of variables associated with the future evolution of GDP, which have been previously filtered to avoid incorporating spurious relationships with the target variable. The weights have been chosen, after analysing several options, ensuring that they minimise the mean square error of the prediction.

After analysing the statistical properties of about 200 economic indicators, 14 variables were identified as those that together can allow a good prediction of GDP up to 6 months ahead. Among these variables are the banking system’s credit to the private sector, the physical volume index of intermediate goods, production and sales of construction bars, the effective reserve requirement rate, central government current account savings, the Support confidence index and the one-year foreign currency lending rate.

As an example, the Leading Indicators system predicts, using data up to March 2002, that the economy in the first half of 2002 will grow at an annual rate of 2.5% and that by the third quarter of the year it will continue to grow at an annual rate of 2.6%.