At any given instance there are many indicators of the state of the economy or the state of a sector and frequently the signals are mixed – some may point to an expansion, others to a contraction, for example. We consider how best to combine this conflicting information into an overall index of economic conditions. This index plays the role of the “underlying cycle” that has the property of minimizing the distortionary impact of noise in the n individual signals. This is essentially the panel regression approach of Stock and Watson (2010). We elaborate and evaluate this rich approach, note its links to stochastic-index-number theory and suggest new interpretations, modifications and extensions. The approach is illustrated with the world prices of six important metals.
