A Reality Check for Data Snooping

By Halbert White. 2000

1,807 citations recorded by the source corpus.

Subfields: Management Science and Operations Research, Signal Processing

View Paper: https://www.semanticscholar.org/paper/ee1014390904db3c2c7c1f81719f49450e621a70

How researchers see this paper

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

In Management Science and Operations Research

In Signal Processing

Citations in the corpus

Copula Methods for Forecasting Multivariate Time Series (2013, methodology): “Below we will focus on pair-wise comparisons of models; for comparisons of large collections of models see White (2000), Romano and Wolf (2005) and Hansen, et al. (2011), for example.”
Comparing Predictive Accuracy, Twenty Years Later: A Personal Perspective on the Use and Abuse of Diebold–Mariano Tests (2015, positive): “…data mining in finite samples, as emphasized by Rossi and Inoue (2012) and Hansen and Timmermann (201 1), because one can then also mine over ť. Achieving robustness to data mining in finite samples requires simulation methods that model the data mining, as in the "reality check" of White (2000).”
Predicting the volatility of the S&P-500 stock index via GARCH models: the role of asymmetries (2005, methodology): “ White (2000) has suggested two ways of obtaining valid critical values for the distribution of maxk=1,..., mZP(k).”
Advances in Forecast Evaluation (2011, positive): “…below.22 Step 1: Relabel the out-of-sample average loss di¤erentials dj from smallest to largest so that ~d1 = mink=1;:::;K dk and ~dK = maxk=1;:::;K dk: Step 2: Use the bootstrap procedure of White (2000) to estimate the critical value c associated with an -level one-shot test as described above.”
Stock Returns on Customer Satisfaction Do Beat the Market: Gauging the Effect of a Marketing Intangible (2016, background): “Stock Returns on Customer Satisfaction / 95 data-driven discovery leads to seriously distorted findings and that it is endemic in the literature (White 2000).”
Impossibility Results for Nondifferentiable Functionals (2012, background): “1The dot notation in κ̇ suggests a derivative, and this will be useful for the asymptotic theory in the next subsection.”
Evaluating Volatility and Correlation Forecasts (2009, background): “When interest lies in comparing many competing forecasts there are two main approaches: the “Reality Check” [62], and a modified version with better power properties [34], or the “Model Confidence Set” [38].”
A Transaction-Cost Perspective on the Multitude of Firm Characteristics (2020, positive): “…only after their publication as in McLean and Pontiff (2016), excluding microcaps, firm size, shortsale constraints, applying the reality check in White (2000), expanding our dataset to also consider characteristics with a large number of missing observations, different subperiods, the…”
Asymptotics for out of sample tests of Granger causality (2007, background): “White (2000) and Hansen (2003) show how to use the bootstrap to compensate for data-snooping biases when comparing the predictive ability of a large number of models.”
Forecasting: theory and practice (2020, methodology): “Addressing a need for a reality check on “data snooping”, White (2000) later modified by Hansen (2005) developed a multiple model test that uses a null hypothesis of “superior predictive ability” instead of the equal predictive ability used in DM tests.”
FORECAST EVALUATION (methodology): “White (2000) recommends the stationary bootstrap of Politis and Romano (1994).”