FORECAST EVALUATION

How researchers see this paper

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

Citations in the corpus

The Tourism Forecasting Competition (2011, background): “There is a gap in the literature between the methodological developments in density forecasting (see, for example, Tay and Wallis, 2002; West, 2006) and the applied forecasting papers that do not produce statements of uncertainty to accompany their point forecasts.”
Higher order elicitability and Osband’s principle (2015, background): “Gneiting (2011) demonstrated impressively that scoring functions should be incentive compatible in that they should encourage the forecasters to issue truthful reports; see also Engelberg, Manski and Williams (2009), Murphy and Daan (1985).”
Copula Methods for Forecasting Multivariate Time Series (2013, background): “This is an important aspect of the evaluation of economic forecasts, see West (2006) for motivation and discussion.”
Equitability Revisited: Why the ''Equitable Threat Score'' Is Not Equitable (2010, background): “Murphy and Daan (1985) recognized this difﬁculty and deﬁned the weaker property ‘‘consistency’’: if a probabilistic belief is converted into a deterministic forecast via some ‘‘direc-tive’’ (e.g., forecast occurrence only if you believe the event has greater than a 50% probability of occurring),…”
Evaluating Probabilistic Forecasts Using Information Theory (2002, background): “…The minimum value of Pred ( E [IGN]) occurs only when g i (cid:53) f i ; thus, ignorance is a strictly proper scoring rule (Lindley 1985; Murphy and Daan 1985; Winkler and Murphy 1968): forecasters cannot expect to reduce their ignorance score by issuing a PDF different from their best judgment .”
Making and Evaluating Point Forecasts (2009, background): “…desirable for administrative or political leadership to provide a directive or target functional to Census Bureau scientists, much in the way that Murphy and Daan (1985) and Engelberg, Manski, and Williams (2009) requested guidance for point forecasters, in the quotes that open and motivate this…”
Extremal Dependence Indices: Improved Verification Measures for Deterministic Forecasts of Rare Binary Events (2011, background): “The notion of consistency (Murphy and Daan 1985) provides another way to deﬁne hedging for deterministic forecasts.”
A skill score based on economic value for probability forecasts (2001, methodology): “Almost universally, the accuracy of such forecasts is evaluated using the Brier Score (e.g., Murphy, 1997; Murphy & Daan, 1985; Stanski et al., 1989; Wilks, 1995): (1) where ft is the probability forecast on occasion t, ot is a binary variable indicating whether the event occurred (ot = 1) or not…”
Averaging Forecasts from Vars with Uncertain Instabilities (2006, unknown): “(2005), for the reasons given in the Corradi and Swanson (2006) and West (2006) surveys of the literature on testing for differences in forecast accuracy.”
Approximately Normal Tests for Equal Predictive Accuracy in Nested Models (2007, methodology): “As in Clark and West (2005), we call this MSPE-adjusted: (3.15) P-1 tT=R ^e21t+1 - [P-1 tT=R ^e22t+1-P-1 tT=R( ^y1t+1^y2t+1) 2 ] ^2 1-( ^2 2-adj.).”
Exchange Rates and Fundamentals: Footloose or Evolving Relationship? (2009, background): “While the price level could be thought of as a function of the monetary fundamentals used in this paper, in ation could move in a di¤erent way than the level and be linked to exchange rate uctuations via central banks reaction functions (e.g., Engel and West, 2006).”
Statistical Analysis in Climate Research (2000, unknown): “For further reading, we also recommend Murphy and Daan [284], Murphy and Winkler [286], Murphy, Brown, and Chen [283], Barnston [25] and Livezey [256].”
Comparing forecast accuracy: A Monte Carlo investigation (2013, background): “…dAD = P̂ −1∑T t=R+h dAD, t, dAD, t = e 2 1t−e22t +(ŷ1t − ŷ2t) 2 and σ̂2AD (m) is the non-parametric estimator of the long run variance of dAD, t, that par- 4It is however argued that the Gaussian critical values would still approximately hold if P/R is small (e.g. less than 0.1, see West, 2006).”
Evaluating the Predictive Accuracy of Volatility Models (2001, methodology): “Just as with the statistical loss functions described in Section II, a variety of proper scoring rules are available; see Murphy and Daan (1985) as well as Winkler (1993) for further discussion.”
On the Forecasting Accuracy of Multivariate GARCH Models (2010, background): “Important generalizations can be found West (1996), Clark and McCracken (2001), Clark and West (2006) and Clark and West (2007).”