Forecast evaluation of small nested model sets / Kirstin Hubrich, Kenneth D. West.
Material type: TextSeries: Working paper series (National Bureau of Economic Research) ; no. 14601.Publication details: Cambridge, Mass. : National Bureau of Economic Research, 2008Description: 24, [9] p. : ill. ; 22 cmSubject(s): Econometric models -- Evaluation | Forecasting -- Statistical methods | Economics -- Statistical methods | Inflation (Finance) -- United States -- Econometric modelsLOC classification: HB1 | .N38 no. 14601Online resources: Click here to access online Summary: We propose two new procedures for comparing the mean squared prediction error (MSPE) of a benchmark model to the MSPEs of a small set of alternative models that nest the benchmark. Our procedures compare the benchmark to all the alternative models simultaneously rather than sequentially, and do not require reestimation of models as part of a bootstrap procedure. Both procedures adjust MSPE differences in accordance with Clark and West (2007); one procedure then examines the maximum t-statistic, the other computes a chi-squared statistic. Our simulations examine the proposed procedures and two existing procedures that do not adjust the MSPE differences: a chi-squared statistic, and White₂s (2000) reality check. In these simulations, the two statistics that adjust MSPE differences have most accurate size, and the procedure that looks at the maximum t-statistic has best power. We illustrate our procedures by comparing forecasts of different models for U.S. inflation.Item type | Current library | Collection | Call number | Copy number | Status | Date due | Barcode |
---|---|---|---|---|---|---|---|
Book | University of Macedonia Library Βιβλιοστάσιο Β (Stack Room B) | Research Papers | HB1.N38 no. 14601 (Browse shelf (Opens below)) | 1 | Available | 0013116457 |
Includes bibliographical references (p. [1-2]).
We propose two new procedures for comparing the mean squared prediction error (MSPE) of a benchmark model to the MSPEs of a small set of alternative models that nest the benchmark. Our procedures compare the benchmark to all the alternative models simultaneously rather than sequentially, and do not require reestimation of models as part of a bootstrap procedure. Both procedures adjust MSPE differences in accordance with Clark and West (2007); one procedure then examines the maximum t-statistic, the other computes a chi-squared statistic. Our simulations examine the proposed procedures and two existing procedures that do not adjust the MSPE differences: a chi-squared statistic, and White₂s (2000) reality check. In these simulations, the two statistics that adjust MSPE differences have most accurate size, and the procedure that looks at the maximum t-statistic has best power. We illustrate our procedures by comparing forecasts of different models for U.S. inflation.
There are no comments on this title.