Smoothing Variance Estimates for Price Indexes Over Time

This paper explores the possibility of developing generalized variances for price indexes by applying nonparametric scatterplot smoothers to time series of point variance estimates. The goal here is to formulate smoothed variances which are approximately unbiased, which provide acceptable confidence interval coverage, and which are more stable than the point variance estimates. Smoothing methods are applied to time series of point variance estimates in a simulation study using data from the U.S. Consumer Price Index program.