A composite leading indicator for the Peruvian economy based on the BCRP's monthly business tendency surveys

This paper documents the construction of a composite leading indicator for the Peruvian economy based on the business tendency surveys (BTS) conducted by the Banco Central de Reserva del Perú (BCRP). We first classify potential composite leading indicators into "semantic" and "sophisti-cated" types. The former are based on the contents of the underlying indicators, whereas the latter results from statistical analyses relating to pre-determined reference series. We show that the BCRP BTS data provides a suitable basis for the construction of a sophisticated indicator with the Peru-vian year-on-year GDP growth rate as a reference series. The indicator selection consists of a num-ber of steps comprising semantic analyses of the questionnaire items, cross-correlation analyses as well as turning point analyses. We argue that based on these analyses, the choice should fall on five indicators, relating to firm-specific questionnaire items as well as to items relating to the sector or economy as a whole. The composite leading indicator is computed as the fist principal component of the selected variables. In-sample, it shows a lead of four months before the reference series, which amounts to about six months before the first official data release dates. Due to the limited number of observations (the BCRP's BTS now covering about eight years), we did not reserve any data points for out-of-sample analyses of the suggested composite leading indicator. Accordingly, the performance of the indicator still has to stand the test of time and its lead should be carefully monitored.