|
on Forecasting |
| By: | Mehmet Balcilar (Department of Economics, Faculty of Business and Economics, Eastern Mediterranean University); Nico Katzke (Department of Economics, Stellenbosch University, South Africa); Rangan Gupta (Department of Economics, University of Pretoria) |
| Abstract: | In this paper we test whether the key metals prices of gold and platinum significantly improve infl ation forecasts for the South African economy. We also test whether controlling for conditional correlations in a dynamic setup, using bivariate Bayesian-Dynamic Conditional Correlation (B-DCC) models, improves infl ation forecasts. To achieve this we compare out-of-sample forecast estimates of the B-DCC model to RandomWalk, Autoregressive and Bayesian VAR models. We find that for both the BVAR and BDCC models, improving point forecasts of the Autoregressive model of in flation remains an elusive exercise. This, we argue, is of less importance relative to the more informative density forecasts. For this we find improved forecasts of infl ation for the B-DCC models at all forecasting horizons tested. We thus conclude that including metals price series as inputs to infl ation models leads to improved density forecasts, while controlling for the dynamic relationship between the included price series and in flation similarly leads to significantly improved density forecasts. |
| Keywords: | Bayesian VAR, Dynamic Conditional Correlation, Density forecasting, Random Walk, Autoregressive model |
| JEL: | C11 C15 E17 |
| Date: | 2015–02 |
| URL: | http://d.repec.org/n?u=RePEc:pre:wpaper:201510&r=for |
| By: | Berg, Tim Oliver |
| Abstract: | In this paper I assess the ability of Bayesian vector autoregressions (BVARs) and dynamic stochastic general equilibrium (DSGE) models of different size to forecast comovements of major macroeconomic series in the euro area. Both approaches are compared to unrestricted VARs in terms of multivariate point and density forecast accuracy measures as well as event probabilities. The evidence suggests that BVARs and DSGE models produce accurate multivariate forecasts even for larger datasets. I also detect that BVARs are well calibrated for most events, while DSGE models are poorly calibrated for some. In sum, I conclude that both are useful tools to achieve parameter dimension reduction. |
| Keywords: | BVARs, DSGE Models, Multivariate Forecasting, Large Dataset, Simulation Methods, Euro Area |
| JEL: | C11 C52 C53 E37 |
| Date: | 2015–02–24 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:62405&r=for |
| By: | Thomas Lux (Department of Economics, University of Kiel, Kiel, Germany); Mawuli Segnon (Department of Economics, University of Kiel, Germany); Rangan Gupta (Department of Economics, University of Pretoria) |
| Abstract: | This paper uses the Markov-switching multifractal (MSM) model and generalized autoregressive conditional heteroscedasticity (GARCH)-type models to forecast oil price volatility over the time periods from January 02, 1875 to December 31, 1895 and from January 03, 1977 to March 24, 2014. Based on six dierent loss functions and by means of the superior predictive ability (SPA) test, we evaluate and compare their forecasting performance at short and long horizons. The empirical results indicate that none of our volatility models can uniformly outperform other models across all six different loss functions. However, the new MSM model comes out as the model that most often across forecasting horizons and subsamples cannot be outperformed by other models, with long memory GARCH-type models coming out second best. |
| Keywords: | Crude oil prices, GARCH, Multifractal processes, SPA test |
| JEL: | C52 C53 C22 |
| Date: | 2015–03 |
| URL: | http://d.repec.org/n?u=RePEc:pre:wpaper:201511&r=for |
| By: | Plambeck, Erica (Stanford University); Bayati, Mohsen (Stanford University); Ang, Erjie (?); Kwasnick, Sara (?); Aratow, Mike (?) |
| Abstract: | This paper proposes a Combined Method (combining fluid model estimators and statistical learning) to forecast the wait time for low-acuity patients in an Emergency Department, and describes the implementation of the Combined Method at the San Mateo Medical Center (SMMC). In historical data from four different hospitals, the Combined Method is more accurate than stand-alone fluid model estimators and statistical learning, and also more accurate than the rolling averages that hospitals currently use to forecast the ED wait time. In historical data and post-implementation data for SMMC, the Combined Method reduces the mean squared forecast error by a nearly third relative to the best rolling average, notably by correcting errors of underestimation in which a patient waits for longer than the forecast. The paper provides a general recipe by which any hospital with an Electronic Medical Records (EMR) can implement the Combined Method. |
| Date: | 2014 |
| URL: | http://d.repec.org/n?u=RePEc:ecl:stabus:3187&r=for |
| By: | Maheu, John M; Yang, Qiao |
| Abstract: | The time-series dynamics of short-term interest rates are important as they are a key input into pricing models of the term structure of interest rates. In this paper we extend popular discrete time short-rate models to include Markov switching of infinite dimension. This is a Bayesian nonparametric model that allows for changes in the unknown conditional distribution over time. Applied to weekly U.S. data we find significant parameter change over time and strong evidence of non-Gaussian conditional distributions. Our new model with an hierarchical prior provides significant improvements in density forecasts as well as point forecasts. We find evidence of recurring regimes as well as structural breaks in the empirical application. |
| Keywords: | hierarchical Dirichlet process prior, beam sampling, Markov switching, MCMC |
| JEL: | C11 C14 C22 C58 |
| Date: | 2015–01 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:62408&r=for |
| By: | Conrad, Christian; Loch, Karin |
| Abstract: | We propose a new measure of the expected variance risk premium that is based on a forecast of the conditional variance from a GARCH-MIDAS model. We find that the new measure has strong predictive ability for future U.S. aggregate stock market returns and rationalize this result by showing that the new measure effectively isolates fundamental uncertainty as the factor that drives the variance risk premium. |
| Keywords: | Variance risk premium; return predictability; VIX; GARCH-MIDAS; economic uncertainty; vol-of-vol |
| Date: | 2015–02–27 |
| URL: | http://d.repec.org/n?u=RePEc:awi:wpaper:0583&r=for |
| By: | Kasznik, Ron (Stanford University); Kremer, Ilan (?) |
| Date: | 2014–06 |
| URL: | http://d.repec.org/n?u=RePEc:ecl:stabus:3046&r=for |