nep-for New Economics Papers
on Forecasting
Issue of 2019‒04‒01
six papers chosen by
Rob J Hyndman
Monash University

  1. Forecasting Financial Stress Indices in Korea: A Factor Model Approach By Hyeongwoo Kim; Wen Shi; Hyun Hak Kim
  2. Nonparametric Predictive Regressions for Stock Return Prediction By Cheng, T.; Gao, J.; Linton, O.
  3. Forecasting daily electricity prices with monthly macroeconomic variables By Foroni, Claudia; Ravazzolo, Francesco; Rossini, Luca
  4. Bayesian MIDAS penalized regressions: estimation, selection, and prediction By Matteo Mogliani
  5. The Impact of Jumps and Leverage in Forecasting the Co-Volatility of Oil and Gold Futures By Asai, M.; Gupta, R.; McAleer, M.J.
  6. Ensemble Methods for Causal Effects in Panel Data Settings By Susan Athey; Mohsen Bayati; Guido Imbens; Zhaonan Qu

  1. By: Hyeongwoo Kim; Wen Shi; Hyun Hak Kim
    Abstract: We propose factor-based out-of-sample forecast models for Korea's financial stress index and its 4 sub-indices that are developed by the Bank of Korea. We extract latent common factors by employing the method of the principal components for a panel of 198 monthly frequency macroeconomic data after differencing them. We augment an autoregressive-type model of the financial stress index with estimated common factors to formulate out-of-sample forecasts of the index. Our models overall outperform both the stationary and the nonstationary benchmark models in forecasting the financial stress indices for up to 12-month forecast horizons. The first common factor that represents not only financial market but also real activity variables seems to play a dominantly important role in predicting the vulnerability in the financial markets in Korea.
    Keywords: Financial Stress Index; Principal Component Analysis; PANIC; In-Sample Fit; Out-of-Sample Forecast; Diebold-Mariano-West Statistic
    JEL: E44 E47 G01 G17
    Date: 2019–03
  2. By: Cheng, T.; Gao, J.; Linton, O.
    Abstract: We propose two new nonparametric predictive models: the multi-step nonparametric predictive regression model and the multi-step additive predictive regression model, in which the predictive variables are locally stationary time series. We define estimation methods and establish the large sample properties of these methods in the short horizon and the long horizon case. We apply our methods to stock return prediction using a number of standard predictors such as dividend yield. The empirical results show that all of these models can substantially outperform the traditional linear predictive regression model in terms of both in-sample and out-of-sample performance. In addition, we _nd that these models can always beat the historical mean model in terms of in-sample fitting, and also for some cases in terms of the out-of-sample forecasting. We also compare our methods with the linear regression and historical mean methods according to an economic metric. In particular, we show how our methods can be used to deliver a trading strategy that beats the buy and hold strategy (and linear regression based alternatives) over our sample period.
    Keywords: Kernel estimator, locally stationary process, series estimator, stock return prediction
    JEL: C14 C22 G17
    Date: 2019–03–25
  3. By: Foroni, Claudia; Ravazzolo, Francesco; Rossini, Luca
    Abstract: We analyse the importance of macroeconomic information, such as industrial production index and oil price, for forecasting daily electricity prices in two of the main European markets, Germany and Italy. We do that by means of mixed-frequency models, introducing a Bayesian approach to reverse unrestricted MIDAS models (RU-MIDAS). We study the forecasting accuracy for different horizons (from 1 day ahead to 28 days ahead) and by considering different specifications of the models. We find gains around 20% at short horizons and around 10% at long horizons. Therefore, it turns out that the macroeconomic low frequency variables are more important for short horizons than for longer horizons. The benchmark is almost never included in the model confidence set. JEL Classification: C11, C53, Q43, Q47
    Keywords: density forecasting, electricity prices, forecasting, MIDAS models, mixed-frequency VAR models
    Date: 2019–03
  4. By: Matteo Mogliani
    Abstract: We propose a new approach to mixed-frequency regressions in a high-dimensional environment that resorts to Group Lasso penalization and Bayesian techniques for estimation and inference. To improve the sparse recovery ability of the model, we also consider a Group Lasso with a spike-and-slab prior. Penalty hyper-parameters governing the model shrinkage are automatically tuned via an adaptive MCMC algorithm. Simulations show that the proposed models have good selection and forecasting performance, even when the design matrix presents high cross-correlation. When applied to U.S. GDP data, the results suggest that financial variables may have some, although limited, short-term predictive content.
    Keywords: MIDAS regressions, penalized regressions, variable selection, forecasting, Bayesian estimation.
    JEL: C11 C22 C53 E37
    Date: 2019
  5. By: Asai, M.; Gupta, R.; McAleer, M.J.
    Abstract: The paper investigates the impact of jumps in forecasting co-volatility in the presence of leverage effects. We modify the jump-robust covariance estimator of Koike (2016), such that the estimated matrix is positive definite. Using this approach, we can disentangle the estimates of the integrated co-volatility matrix and jump variations from the quadratic covariation matrix. Empirical results for daily crude oil and gold futures show that the co-jumps of the two futures have significant impacts on future co-volatility, but that the impact is negligible in forecasting weekly and monthly horizons
    Keywords: Commodity Markets, Co-volatility, Forecasting, Jump, Leverage Effects, Realized, Covariance, Threshold Estimation.
    JEL: C32 C33 C58 Q02
    Date: 2019–03–01
  6. By: Susan Athey; Mohsen Bayati; Guido Imbens; Zhaonan Qu
    Abstract: In many prediction problems researchers have found that combinations of prediction methods (“ensembles”) perform better than individual methods. A simple example is random forests, which combines predictions from many regression trees. A striking, and substantially more complex, example is the Netflix Prize competition where the winning entry combined predictions using a wide variety of conceptually very different models. In macro-economic forecasting researchers have often found that averaging predictions from different models leads to more accurate forecasts. In this paper we apply these ideas to synthetic control type problems in panel data setting. In this setting a number of conceptually quite different methods have been developed, with some assuming correlations between units that are stable over time, others assuming stable time series patterns common to all units, and others using factor models. With data on state level GDP for 270 quarters, we focus on three basic approaches to predicting missing values, one from each of these strands of the literature. Rather than try to test the different models against each other and find a true model, we focus on combining predictions based on each of the separate models using ensemble methods. For the ensemble predictor we focus on a weighted average of the three individual methods, with non-negative weights determined through out-of-sample cross-validation.
    JEL: C01 C14
    Date: 2019–03

This nep-for issue is ©2019 by Rob J Hyndman. It is provided as is without any express or implied warranty. It may be freely redistributed in whole or in part for any purpose. If distributed in part, please include this notice.
General information on the NEP project can be found at For comments please write to the director of NEP, Marco Novarese at <>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the School of Economics and Finance of Massey University in New Zealand.