> > >
GDPplus is a measure of the quarter-over-quarter rate of growth of real GDP in annualized percentage points. It improves on the BEA's expenditure- and income-side measures, GDP_E and GDP_I, respectively. GDP_E is the "standard" GDP measure used routinely, whereas GDP_I is little used, but each contains useful information. In particular, as proposed in Aruoba, Diebold, Nalewaik, Schorfheide, and Song (ADNSS)
, one can view both GDP_E and GDP_I as noisy indicators of underlying latent true GDP, which can then be extracted using optimal filtering methods. Here we implement the ADNSS framework. We call the optimal GDP extraction GDPplus. On this web page, we provide GDPplus data. We will update the page whenever a new GDPplus vintage becomes available, whether due to a new release or a revision of the underlying GDP_E and GDP_I indicators. Note that all annualized growth rates use the formula for continuous compounding. We estimate the ADNSS two-equation model with the restriction on the variance matrix for the shocks. Estimation uses the method of maximum likelihood.
The GDPplus methodology is based on the measurement-error approach discussed in the following papers:
Aruoba, S.B., F.X. Diebold, J. Nalewaik, F. Schorfheide, and D. Song, "Improving GDP Measurement: A Measurement-Error Perspective ," FRB Philadelphia Working Paper 13-16 (May 2013).
Aruoba, S.B., F.X. Diebold, J. Nalewaik, F. Schorfheide, and D. Song, "Improving GDP Measurement: A Measurement-Error Perspective," Journal of Econometrics, 191 (2016), pp. 384-397.
Additional references are provided in the above papers.