Benchmarking through calibration of weights for microdata

Benchmarking of subannual to annual series in official statistics is usually carried out using a macrolevel procedure. This ensures that subannual estimates are consistent with annual totals, with some appropriate methods to ensure continuity of the series between years. Another approach would be to use a micro-level procedure. This procedure would use the annual totals as auxiliary data in calculating survey weights, thereby adjusting the subannual weights so that the annual and subannual estimates are consistent. Some continuity adjustments are also required with this method to give a plausible series across years. The correspondence and differences between the two approaches are illustrated using a simple simulation example. The benefits and drawbacks of the two different approaches are also discussed.