Statistical inference on series of atmospheric chemistry data

This study is motivated by a problem concerning comparison of atmospheric chemistry data obtained from two different measurement networks. The data describe pollutant concentrations in air. One network analyzes gases and particles captured on filters daily, the other weekly. The networks use similar but not identical instrumentation and measurement methods. Chemical concentrations form usually non-stationary time series. The goal of the paper is to present a multivariate linear Gaussian model with time dependent correlation matrix for description of such series, a method for parameter estimation and fitting of the model and the procedure used for inference about the mentioned data.