A Review of Seasonal Adjustment Diagnostics

Seasonal adjustment methods are used to process and publish thousands of time series across the world each month, judgement adequacy relies heavily upon seasonal diagnostics. This paper discusses tests for a adjustment, first reviewing broader background on seasonality then proceeding four that appropriate stationary forms seasonality. We contrast time-domain frequency-domain approaches, focusing diagnostics available in literature (and software packages) applying large collection public use series. Each is designed around distinct formulation seasonality, hence, empirical performances differ; we compare include discussion how diagnostic results can be improve faulty adjustments. Directions future research, involving inverse partial autocorrelations polyspectra, also discussed, methodologies supported by theoretical simulation results.