Approximately Normal Tests for Equal Predictive Accuracy in Nested Models

By Todd E. Clark, Ken West. 2007

2,234 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Finance, Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/02f08822c3d5d124d8c02a7172f4ea37543d3c94

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

In Finance

In Management Science and Operations Research

Citations in the corpus

Statistical tests for multiple forecast comparison (2012, background): “See, among others, Giacomini and White (2006), Clark and West (2006), Clark and West (2007) and McCracken (2007).”
Constructing a Global Fear Index for the COVID-19 Pandemic (2020, methodology): “The C-T stat denotes the Campbell-Thompson (2008) test while the C-W test is the Clark and West (2007) test.”
Uncertainty Due to Infectious Diseases and Energy Market Volatility (2020, methodology): “Since the two models are nested as the historical average is a restricted version of Equation (1), their forecast performance comparison can easily be conducted using the Clark and West (2007) [CW] test.”
Some Theoretical Results on Forecast Combinations (2018, positive): “Their argument is based on obtaining weights that minimises the MSFE and rely on the fact that the general weighting schemes nest the simple average, which can be extended to the forecast evaluation of competing (nested) models as in Clark and West (2007).”
Comparing forecast accuracy: A Monte Carlo investigation (2013, background): “Since there is no additional temporal dependence to be taken into account, we calculate the statistics (2), (5), (6), and (8) for m = 0, i.”
FORECAST EVALUATION (methodology): “Across 48 sets of simulations, with DGPs calibrated to match key characteristics of asset price data, Clark and West (2005b) found that the median size of a nominal 10% test using the standard result (6.8) was less than 1%.”
Out-of-Sample Forecast Tests Robust to the Choice of Window Size (2011, positive): “Appendix 1 shows that Proposition 1 applies to the test statistic (13) under broad conditions, which are similar to those discussed for eq.”
Advances in Forecast Evaluation (2011, positive): “For the ENC-t test applied to nested forecasting models, Clark and West (2006, 2007) show that, under certain conditions, the distribution is either asymptotically normal or approximately normal in practice.”
Complete subset regressions (2013, methodology): “To compare model performance more formally, we use the test proposed by Clark and West (2007), treating the simple prevailingmean forecast as our benchmark.”
Pandemics and the Asia-Pacific Islamic Stocks (2020, methodology): “We use the historical average (constant returns) model that ignores the UPE index as the baseline model, and its relative performance with the UPE-based models (“Without control” and “With control”) is evaluated using the Clark & West (2007) test.”