Denton’s Temporal Disaggregation Methods with the Application of Bangladesh’s Annual Gross Domestic Product for Quarterly Benchmarking and Forecasting
Different methods of temporal disaggregation are discussed in detail; mainly the methods developed by Denton in 1971 and other purely mathematical methods. First, the original method developed by Denton and its solution are described by referencing Denton’s original article. The Cholette–Dagum regression-based method (or Denton-Cholette method) is also included to enrich the comparison. Bangladesh’s annual export figures are then disaggregated into a quarterly series by Denton’s additive and proportional (first and second difference) methods, and the Denton-Cholette additive and proportional (first and second difference) methods by using R package “tempdisagg”. The quarterly imports of capital goods and others (Iron, steel and other base metals; and capital machinery) in Bangladesh are used as the indicator series for the fiscal years FY2009 to FY2019. By comparing the estimated series with the real quarterly exports series, with the aid of root mean squared errors, it is concluded that the Denton-Cholette additive method (the first difference) performs better than the Denton-Cholette’s proportional variants as well as Denton’s additive and proportional variants.