Growth Cycles in XXth Century European Industrial Productivity: Unbiased Variance Estimation in a Time-varying Parameter Model

This note applies the median unbiased estimation of coefficient variance, proposed by Stock and Watson (1998), to the extraction of the time-varying trend growth rate of industrial productivity in fifteen European countries, over most of the XXth Century, by means of an unobservable components univariate decomposition. In addition to the description of the procedure, this illustration is particularly useful in explaining why the method is especially appropriate for comparison of trends growth rates extracted from time series with diverse degrees of variability.