Optimal Forecast Combination Under General Loss Functions and Forecast Error Distributions

By Graham Elliott, Allan G. Timmermann. 2002

195 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/c1656aed673f9dadb2f85d0053a8ac12d3cef67c

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

Improving Forecast Accuracy by Combining Recursive and Rolling Forecasts (2008, unknown): “Elliott and Timmermann, 2004) and empirical (e.g. Stock and Watson, 2003, 2004). Second, our analysis follows very much in the spirit of Min and Zellner (1993), who also consider forecast combination as a means of handling heterogeneity induced by structural change. Using a Bayesian framework, they ”
Properties of Optimal Forecasts under Asymmetric Loss and Nonlinearity (2007, methodology): “We next provide new analytical results under linearexponential (linex) loss and a Markov switching DGP. Linex loss is frequently used in the literature on asymmetric loss, c.f. Batchelor and Peel (1998), Varian (1974), Zellner (1986), Sancetta and Satchell (2004) and Elliott and Timmermann (2004).”
Bias in Federal Reserve Inflation Forecasts: Is the Federal Reserve Irrational or Just Cautious? (2005, unknown): “Equation (22) is satisfied if the forecasts are optimal, if they were produced using an asymmetric quadratic loss function with parameter φ, if vt is in the information set of the producer of the forecasts, and if δ = 0.”
Markov Switching Models in Empirical Finance (2011, background): “Elliott and Timmermann (2004) have argued that a combination scheme that puts more weight on adaptive (e.”
Forecast Combination and Model Averaging Using Predictive Measures (2007, unknown): “While the literature on forecast combination is extensive, see Clemen (1989) for a somewhat dated review, and Hendry and Clements (2004), and Elliott and Timmermann (2004) for recent theoretical contributions, relatively little attention has been given to the use of predictive measures of fit as…”
Forecasting with temporal hierarchies (2017, methodology): “…been widely investigated, resulting in various sophisticated weighting methods (see Newbold and Granger, 1974; Min and Zellner, 1993; Elliott and Timmermann, 2004, 2005); yet, simple approaches, such as the unweighted average, are found to perform as well as more sophisticated approaches…”
Persistence in forecasting performance and conditional combination strategies (2006, background): “However, given a large number of forecasting models (N) relative to the number of time-series 7Including a constant to capture bias effects is a strategy recommended (under MSE loss) by Granger and Ramanathan (1984) and, for a variety of loss functions, by Elliott and Timmermann (2004).”
FORECAST EVALUATION (background): “…include MSPE, mean prediction error, mean absolute prediction error, covariance between one model's prediction and another model's prediction error, mean utility or profit, and means of loss functions that weight positive and negative errors asymmetrically as in Elliott and Timmermann (2003).”
Explanations of the inconsistencies in survey respondents'forecasts (2010, unknown): “, Granger, White and Kamstra (1989), Christo¤ersen (1998) and Giacomini and Komunjer (2005)), volatility forecasts (e.”
Nonparametric Bootstrap Procedures for Predictive Inference Based on Recursive Estimation Schemes (2007, background): “Clark and McCracken (2005) extend McCracken (2004) to the case of multistep ahead forecasts. ability of a fixed number of nested models have previously also been suggested.”
Forecast combinations for value at risk and expected shortfall (2020, background): “Elliott and Timmermann (2004) show that the question of whether combining weights should be equal depends on the loss function, implying that empirical evidence from the literature on”