In this paper we assess the merits of financial condition indices constructed using simple averages versus a more sophisticated alternative that uses factor models with time varying parameters. Our analysis is based on data for 18 advanced and emerging economies at a monthly frequency covering about 70% of the world's GDP. We use four criteria to assess the performance of these indicators, namely quantile regressions, Structural Vector Autoregressions, the ability of the indices to predict banking crises and their response to US monetary policy shocks. We find that averaging across the indicators of interest, using judgemental but intuitive weights, produces financial condition indices that are not inferior to, and actually perform better than, those constructed with more sophisticated statistical methods.
financial conditions quantile regressions banking crises SVARs spillovers