nep-for New Economics Papers
on Forecasting
Issue of 2021‒02‒01
ten papers chosen by
Rob J Hyndman
Monash University

  1. Mining for Oil Forecasts By ; Nida Cakir Melek; Charles W. Calomiris
  2. Long-term prediction intervals with many covariates By Sayar Karmakar; Marek Chudy; Wei Biao Wu
  3. Split-then-Combine simplex combination and selection of forecasters By Antonio Martin Arroyo; Aranzazu de Juan Fernandez
  4. Measuring Uncertainty of a Combined Forecast and Some Tests for Forecaster Heterogeneity By Kajal Lahiri; Huaming Peng; Xuguang Sheng
  5. Forecasting expected and unexpected losses By Mikael Juselius; Nikola Tarashev
  6. Real-time Inflation Forecasting Using Non-linear Dimension Reduction Techniques By Niko Hauzenberger; Florian Huber; Karin Klieber
  7. Estimating real-world probabilities: A forward-looking behavioral framework By Ricardo Cris\'ostomo
  8. On the Aggregation of Probability Assessments: Regularized Mixtures of Predictive Densities for Eurozone Inflation and Real Interest Rates By Francis X. Diebold; Minchul Shin; Boyuan Zhang
  9. Text-based recession probabilities By Ferrari, Massimo; Le Mezo, Helena
  10. Global Smartphones Sales May Have Peaked By Joannes Mongardini; Aneta Radzikowski

  1. By: ; Nida Cakir Melek; Charles W. Calomiris
    Abstract: In this paper, we study the usefulness of a large number of traditional determinants and novel text-based variables for in-sample and out-of-sample forecasting of oil spot and futures returns, energy company stock returns, oil price volatility, oil production, and oil inventories. After carefully controlling for small-sample biases, we find compelling evidence of in-sample predictability. Our text measures hold their own against traditional variables for oil forecasting. However, none of this translates to out-of-sample predictability until we data mine our set of predictive variables. Our study highlights that it is difficult to forecast oil market outcomes robustly.
    Keywords: Asset Pricing; Commodity Markets; Energy Forecasting; Model Validation
    JEL: C52 G18 G14 G17 Q47
    Date: 2020–12–23
  2. By: Sayar Karmakar; Marek Chudy; Wei Biao Wu
    Abstract: Accurate forecasting is one of the fundamental focus in the literature of econometric time-series. Often practitioners and policy makers want to predict outcomes of an entire time horizon in the future instead of just a single $k$-step ahead prediction. These series, apart from their own possible non-linear dependence, are often also influenced by many external predictors. In this paper, we construct prediction intervals of time-aggregated forecasts in a high-dimensional regression setting. Our approach is based on quantiles of residuals obtained by the popular LASSO routine. We allow for general heavy-tailed, long-memory, and nonlinear stationary error process and stochastic predictors. Through a series of systematically arranged consistency results we provide theoretical guarantees of our proposed quantile-based method in all of these scenarios. After validating our approach using simulations we also propose a novel bootstrap based method that can boost the coverage of the theoretical intervals. Finally analyzing the EPEX Spot data, we construct prediction intervals for hourly electricity prices over horizons spanning 17 weeks and contrast them to selected Bayesian and bootstrap interval forecasts.
    Date: 2020–12
  3. By: Antonio Martin Arroyo; Aranzazu de Juan Fernandez
    Abstract: This paper considers the Split-Then-Combine (STC) approach (Arroyo and de Juan, 2014) to combine forecasts inside the simplex space, the sample space of positive weights adding up to one. As it turns out, the simplicial statistic given by the center of the simplex compares favorably against the fixed-weight, average forecast. Besides, we also develop a Combine-After-Selection (CAS) method to get rid of redundant forecasters. We apply these two approaches to make out-of-sample one-step ahead combinations and subcombinations of forecasts for several economic variables. This methodology is particularly useful when the sample size is smaller than the number of forecasts, a case where other methods (e.g., Least Squares (LS) or Principal Component Analysis (PCA)) are not applicable.
    Date: 2020–12
  4. By: Kajal Lahiri; Huaming Peng; Xuguang Sheng
    Abstract: We have argued that from the standpoint of a policy maker who has access to a number of expert forecasts, the uncertainty of a combined forecast should be interpreted as that of a typical forecaster randomly drawn from the pool. With a standard factor decomposition of a panel of forecasts, we show that the uncertainty of a typical forecaster can be expressed as the disagreement among the forecasters plus the volatility of the common shock. Using new statistics to test for the homogeneity of idiosyncratic errors under the joint limits with both T and n approaching infinity simultaneously, we find that some previously used measures significantly underestimate the conceptually correct benchmark forecast uncertainty.
    Keywords: disagreement, forecast combination, panel data, uncertainty
    JEL: C12 C33 E37
    Date: 2020
  5. By: Mikael Juselius; Nikola Tarashev
    Abstract: Extending a standard credit-risk model illustrates that a single factor can drive both expected losses and the extent to which they may be exceeded in extreme scenarios, ie "unexpected losses". This leads us to develop a framework for forecasting these losses jointly. In an application to quarterly US data on loan charge-offs from 1985 to 2019, we find that financial-cycle indicators – notably, the debt service ratio and credit-to-GDP gap – deliver reliable real-time forecasts, signalling turning points up to three years in advance. Provisions and capital that reflect such forecasts would help reduce the procyclicality of banks' loss-absorbing resources.
    Keywords: loss rate forecasts, cyclical turning points, expected loss provisioning, bank capital
    JEL: G17 G21 G28
    Date: 2020–12
  6. By: Niko Hauzenberger; Florian Huber; Karin Klieber
    Abstract: In this paper, we assess whether using non-linear dimension reduction techniques pays off for forecasting inflation in real-time. Several recent methods from the machine learning literature are adopted to map a large dimensional dataset into a lower dimensional set of latent factors. We model the relationship between inflation and these latent factors using state-of-the-art time-varying parameter (TVP) regressions with shrinkage priors. Using monthly real-time data for the US, our results suggest that adding such non-linearities yields forecasts that are on average highly competitive to ones obtained from methods using linear dimension reduction techniques. Zooming into model performance over time moreover reveals that controlling for non-linear relations in the data is of particular importance during recessionary episodes of the business cycle.
    Date: 2020–12
  7. By: Ricardo Cris\'ostomo
    Abstract: We show that disentangling sentiment-induced biases from fundamental expectations significantly improves the accuracy and consistency of probabilistic forecasts. Using data from 1994 to 2017, we analyze 15 stochastic models and risk-preference combinations and in all possible cases a simple behavioral transformation delivers substantial forecast gains. Our results are robust across different evaluation methods, risk-preference hypotheses and sentiment calibrations, demonstrating that behavioral effects can be effectively used to forecast asset prices. Further analyses confirm that our real-world densities outperform densities recalibrated to avoid past mistakes and improve predictive models where risk aversion is dynamically estimated from option prices.
    Date: 2020–12
  8. By: Francis X. Diebold; Minchul Shin; Boyuan Zhang
    Abstract: We propose methods for constructing regularized mixtures of density forecasts. We explore a variety of objectives and regularization penalties, and we use them in a substantive exploration of Eurozone inflation and real interest rate density forecasts. All individual inflation forecasters (even the ex post best forecaster) are outperformed by our regularized mixtures. The log scores of the Simplex and Best-Average mixtures, for example, are approximately 7% better than that of the ex post best individual forecaster, and 15% better than that of the median forecaster. From the Great Recession onward, the optimal regularization tends to move density forecasts' probability mass from the centers to the tails, correcting for overconfidence.
    Date: 2020–12
  9. By: Ferrari, Massimo; Le Mezo, Helena
    Abstract: This paper proposes a new methodology based on textual analysis to forecast U.S. recessions. Specifically, the paper develops an index in the spirit of Baker et al. (2016) and Caldara and Iacoviello (2018) which tracks developments in U.S. real activity. When used in a standard recession probability model, the index outperforms the yield curve based forecast, a standard method to forecast recessions, at medium horizons, up to 8 months. Moreover, the index contains information not included in yield data that are useful to understand recession episodes. When included as an additional control to the slope of the yield curve, it improves the forecast accuracy by 5% to 30% depending on the horizon. These results are stable to a number of different robustness checks, including changes to the estimation method, the definition of recessions and controlling for asset purchases by major central banks. Yield and textual analysis data also outperform other popular leading indicators for the U.S. business cycle such as PMIs, consumers' surveys or employment data. JEL Classification: E17, E47, E37, C25, C53
    Keywords: forecast, textual analysis, U.S. recessions
    Date: 2021–01
  10. By: Joannes Mongardini; Aneta Radzikowski
    Abstract: Global smartphone sales may have peaked. After reaching nearly 1.5 billion units in 2016, global smartphone sales have since declined, contributing negatively to world trade in 2019 and suggesting that the global market may now be saturated. This paper develops a simple model to forecast smartphone sales, which shows that sales are likely to decline further. As tech companies shift to embedded services (cloud computing, content subscriptions, and financial services), the impact on global trade may also be shifting in favor of services exports mostly from advanced economies.
    Keywords: Mobile internet;Service exports;Stocks;Financial services;Exports;WP,iPhone,Apple,iPhone sale,trade
    Date: 2020–05–29

This nep-for issue is ©2021 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.