Model uncertainty, thick modelling and the predictability of stock returns

Aiolfi, Marco; Favero, Carlo A.

doi:10.1002/for.958

Cited by 93 publications

(37 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, considering the MAPE indicator in Figure 4, the best performing combination for the least volatile load period 6 and for the peak load period 38 is obtained with the models TVR, MS and ARMAX. This agrees with previous research: it has been argued that, rather than combining the full set of forecasts, it is often advantageous to discard the models with the worst performance (see, for instance, Aiolfi and Favero, 2005;Granger and Jeon, 2004;Marcellino, 2004;Watson, 2001, 2004). However, in our study some exceptions emerge when the worst predictive model is the TVR.…”

Section: Ex Post Analysessupporting

confidence: 91%

Combining day-ahead forecasts for British electricity prices

et al. 2013

View full text Add to dashboard Cite

This paper considers how well the approach of combining forecasts extends to the context of electricity prices. With the increasing popularity of regime switching and time-varying parameter models for predicting power prices, the multi model and evolutionary considerations that usually support the combining of simpler time series methods may be less applicable when the individual models incorporate these features. We address this question with a backtesting analysis on British day-ahead prices. Furthermore, given the volatility of power prices and concerns about accurate forecasting under extreme price excursions, we evaluate the results using various error metrics including expected shortfall. The comparisons are furthermore carefully simulated to consider model selection uncertainty in order to realistically test the value of combining as an ex ante policy. Overall, our results support combining for both accurate operational planning and risk management.

show abstract

Section: Ex Post Analysessupporting

confidence: 91%

Combining day-ahead forecasts for British electricity prices

et al. 2013

View full text Add to dashboard Cite

show abstract

“…Most prominently, Stock and Watson (2002) advocate summarizing large panels of predictor variables into a small number of principal components, which are then used for forecasting purposes in a dynamic factor model. Alternative approaches include combining forecasts based on multiple models, each including only a small number of variables (Faust and Wright, 2009;Wright, 2009;Aiolfi and Favero, 2005;Huang and Lee, 2010;Rapach et al, 2010), partial least squares (Groen and Kapetanios, 2008), and Bayesian regression (De Mol et al, 2008;Bańbura et al, 2010;Carriero et al, 2011). Stock and Watson (2009) find that for forecasting macroeconomic time series, the dynamic factor model approach is preferable to these alternatives; see also Ng (2007, 2009) and Ç akmaklı and van Dijk (2010) for successful applications in finance.…”

Section: Introductionmentioning

confidence: 99%

Nonlinear Forecasting with Many Predictors Using Kernel Ridge Regression

et al. 2011

View full text Add to dashboard Cite

This paper puts forward kernel ridge regression as an approach for forecasting with many predictors that are related nonlinearly to the target variable. In kernel ridge regression, the observed predictor variables are mapped nonlinearly into a high-dimensional space, where estimation of the predictive regression model is based on a shrinkage estimator to avoid overfitting. We extend the kernel ridge regression methodology to enable its use for economic time-series forecasting, by including lags of the dependent variable or other individual variables as predictors, as typically desired in macroeconomic and financial applications. Monte Carlo simulations as well as an empirical application to various key measures of real economic activity confirm that kernel ridge regression can produce more accurate forecasts than traditional linear methods for dealing with many predictors based on principal component regression.

show abstract

“…Trimming or filtering out poor forecasters (or models) who mostly contribute noise has been shown to improve forecast combinations (e.g., Aiolfi and Favero (2005)). 9…”

Section: Filtering the Zew Surveymentioning

confidence: 99%

Forecaster Overconfidence and Market Survey Performance

2015

View full text Add to dashboard Cite

Die Dis cus si on Pape rs die nen einer mög lichst schnel len Ver brei tung von neue ren For schungs arbei ten des ZEW. Die Bei trä ge lie gen in allei ni ger Ver ant wor tung der Auto ren und stel len nicht not wen di ger wei se die Mei nung des ZEW dar.Dis cus si on Papers are inten ded to make results of ZEW research prompt ly avai la ble to other eco no mists in order to encou ra ge dis cus si on and sug gesti ons for revi si ons. The aut hors are sole ly respon si ble for the con tents which do not neces sa ri ly repre sent the opi ni on of the ZEW. Forecaster overconfidence and market survey performanceRichard Deaves a Jin Lei b Michael Schröder c ABSTRACTWe document using the ZEW panel of German stock market forecasters that weak forecasters tend to be overconfident in the sense that they provide extreme forecasts and their confidence intervals are less likely to contain eventual realizations. Moderate filters based on forecast accuracy over short rolling windows are somewhat successful in improving predictability. While poor performance can be due to various factors, a filter based on a prior tendency to provide extreme forecasts also improves predictability.

show abstract

Model uncertainty, thick modelling and the predictability of stock returns

Cited by 93 publications

References 54 publications

Combining day-ahead forecasts for British electricity prices

Combining day-ahead forecasts for British electricity prices

Nonlinear Forecasting with Many Predictors Using Kernel Ridge Regression

Forecaster Overconfidence and Market Survey Performance

Contact Info

Product

Resources

About