Estimating an inflation index by quantile regression

Abstract

This article gives a methodology for estimating an inflation index using the quantile regression of Bassett and Koenker. The regression – orthogonal in the logarithmic price changes – is computed by linear programming for each percentile of inflation; and the results are bootstrapped to estimate standard errors. The procedure is applied to monthly data on seven metals.

Keywords:

• inflation index
• quantile regression
• orthogonal regression
• bootstrap

JEL Classification:

• C43
• E31