nep-for New Economics Papers
on Forecasting
Issue of 2020‒02‒24
eleven papers chosen by
Rob J Hyndman
Monash University

  1. An age-at-death distribution approach to forecast cohort mortality By Ugofilippo Basellini; Søren Kjærgaard; Carlo Giovanni Camarda
  2. Where Should We Go? Internet Searches and Tourist Arrivals By Serhan Cevik
  3. Hyperparameter Optimization for Forecasting Stock Returns By Sang Il Lee
  4. Forecasting the Results of Experiments: Piloting an Elicitation Strategy By Stefano DellaVigna; Nicholas Otis; Eva Vivalt
  5. Forecast performance in the ECB SPF: ability or chance? By Meyler, Aidan
  6. Should central banks communicate uncertainty in their projections? By Ryan Rholes; Luba Petersen
  7. Robustness and the general dynamic factor model with infinite-dimensional space: identification, estimation, and forecasting By Trucíos Maza, Carlos César; Mazzeu, João H. G.; Hotta, Luiz Koodi; Pereira, Pedro L. Valls; Hallin, Marc
  8. Flexible (panel) regression models for bivariate count-continuous data with an insurance application By Yang Lu
  9. Improving Oil Price Forecasts by Sparse VAR Methods By Krüger, Jens; Ruths Sion, Sebastian
  10. Semi-Structural Forecasting Model By Francisco Arroyo Marioli; Francisco Bullano; Jorge Fornero; Roberto Zúñiga
  11. Intelligent, Predictive Parking Assist for Trucks By Ioannou, Petros; Monteiro, Fernando Valladares

  1. By: Ugofilippo Basellini; Søren Kjærgaard; Carlo Giovanni Camarda
    Abstract: Mortality forecasting has received increasing interest during recent decades due to the negative financial effects of continuous longevity improvements on public and private institutions’ liabilities. However, little attention has been paid to forecasting mortality from a cohort perspective. In this article, we introduce a novel methodology to forecast adult cohort mortality from age-at-death distributions. We propose a relational model that associates a time-invariant standard to a series of fully and partially observed distributions. Relation is achieved via a transformation of the age-axis. We show that cohort forecasts can improve our understanding of mortality developments by capturing distinct cohort effects, which might be overlooked by a conventional age-period perspective. Moreover, mortality experiences of partially observed cohorts are routinely completed. We illustrate our methodology on adult female mortality for cohorts born between 1835 and 1970 in two high-longevity countries using data from the Human Mortality Database.
    Keywords: Mortality forecasting, Mortality modelling, Relational models, Cohort life table, Smoothing
    Date: 2020–01–26
  2. By: Serhan Cevik
    Abstract: The widespread availability of internet search data is a new source of high-frequency information that can potentially improve the precision of macroeconomic forecasting, especially in areas with data constraints. This paper investigates whether travel-related online search queries enhance accuracy in the forecasting of tourist arrivals to The Bahamas from the U.S. The results indicate that the forecast model incorporating internet search data provides additional information about tourist flows over a univariate approach using the traditional autoregressive integrated moving average (ARIMA) model and multivariate models with macroeconomic indicators. The Google Trends-augmented model improves predictability of tourist arrivals by about 30 percent compared to the benchmark ARIMA model and more than 20 percent compared to the model extended only with income and relative prices.
    Date: 2020–01–31
  3. By: Sang Il Lee
    Abstract: In recent years, hyperparameter optimization (HPO) has become an increasingly important issue in the field of machine learning for the development of more accurate forecasting models. In this study, we explore the potential of HPO in modeling stock returns using a deep neural network (DNN). The potential of this approach was evaluated using technical indicators and fundamentals examined based on the effect the regularization of dropouts and batch normalization for all input data. We found that the model using technical indicators and dropout regularization significantly outperforms three other models, showing a positive predictability of 0.53% in-sample and 1.11% out-of-sample, thereby indicating the possibility of beating the historical average. We also demonstrate the stability of the model in terms of the changes in its feature importance over time.
    Date: 2020–01
  4. By: Stefano DellaVigna; Nicholas Otis; Eva Vivalt
    Abstract: Forecasts of experimental results can clarify the interpretation of research results, mitigate publication bias, and improve experimental designs. We collect forecasts of the results of three Registered Reports preliminarily accepted to the Journal of Development Economics, randomly varying four features: (1) small versus large reference values; (2) whether predictions are in raw units or standard deviations; (3) text-entry versus slider responses; and (4) small versus large slider bounds. Forecasts are generally robust to elicitation features, though wider slider bounds are associated with higher forecasts throughout the forecast distribution. We make preliminary recommendations on how many forecasts should be gathered.
    JEL: O1 O17
    Date: 2020–01
  5. By: Meyler, Aidan
    Abstract: In this paper, we consider whether differences in the forecast performance of ECB SPF respondents reflect ability or chance. Although differences in performance metrics sometimes appear substantial, it is challenging to determine whether they reflect ex ante skill or other factors impacting ex post sampling variation such as the nature of economic shocks that materialised or simply which rounds participants responded in. We apply and adapt an approach developed by D’Agostino et al. (2012) who used US SPF data. They developed a test of a null hypothesis that all forecasters have equal ability. Their statistic reflects both the absolute and relative performance of each forecaster and they used bootstrap techniques to compare the empirical results with the equivalents obtained under the null hypothesis of equal forecaster ability. Our results, at a first pass, suggest that there would appear to be evidence of good/bad forecasters. However once we control for the autocorrelation that is caused by the overlapping rolling horizons, we find, like D’Agostino et al. (2012), that the best forecasters are not statistically significantly better than others. Unlike D’Agostino et al. (2012), however, we do not find evidence of forecasters that perform very significantly worse than others. Controlling for autocorrelation is a key feature of this paper relative to previous work. Our results hold considering the whole sample period of the ECB SPF (1999-2018) as well as the pre- and post-global financial crisis samples. We also find that when assessed across all variables and horizons, the aggregate (consensus) SPF forecast performs best. JEL Classification: C53, E27, E37
    Keywords: bootstrap, forecasting, performance
    Date: 2020–02
  6. By: Ryan Rholes (Texas A&M University); Luba Petersen (Simon Fraser University)
    Abstract: This paper provides original empirical evidence on the emerging practice by central banks of communicating uncertainty in their inflation projections. We compare the effects of point and density projections in a learning-to-forecast laboratory experiment where participants' aggregated expectations about one- and two-period-ahead inflation influence macroeconomic dynamics. Precise point projections are more effective at managing inflation expectations. Point projections reduce disagreement and uncertainty while nudging participants to forecast rationally. Supplementing the point projection with a density forecast mutes many of these benefits. Relative to a point projection, density forecasts lead to larger forecast errors, greater uncertainty about own forecasts, and less credibility in the central bank's projections. We also explore expectation formation in individual-choice environments to understand the motives for responding to projections. Credibility in the projections is significantly lower when strategic considerations are absent, suggesting that projections are primarily effective as a coordination device. Overall, our results suggest that communicating uncertainty through density projections reduces the ecacy of inflation point projections.
    Keywords: expectations, monetary policy, inflation communication, credibility, laboratory experiment, experimental macroeconomics, uncertainty, strategic, coordination, group versus individual choice
    Date: 2020–01
  7. By: Trucíos Maza, Carlos César; Mazzeu, João H. G.; Hotta, Luiz Koodi; Pereira, Pedro L. Valls; Hallin, Marc
    Abstract: General dynamic factor models have demonstrated their capacity to circumvent the curse of dimensionality in time series and have been successfully applied in many economic and financial applications. However, their performance in the presence of outliers has not been analysed yet. In this paper, we study the impact of additive outliers on the identification, estimation and forecasting performance of general dynamic factor models. Based on our findings, we propose robust identification, estimation and forecasting procedures. Our proposal is evaluated via Monte Carlo experiments and in empirical data.
    Date: 2020–02
  8. By: Yang Lu (AMSE - Aix-Marseille Sciences Economiques - EHESS - École des hautes études en sciences sociales - AMU - Aix Marseille Université - ECM - Ecole Centrale de Marseille - CNRS - Centre National de la Recherche Scientifique)
    Abstract: We propose a flexible regression model that is suitable for mixed count-continuous panel data. The model is based on a compound Poisson representation of the continuous variable , with bivariate random effect following a polynomial-expansion-based joint density. Besides the distributional flexibility that it offers, the model allows for closed form forecast updating formu-lae.This property is especially important for insurance applications, in which the future individual insurance premium should be regularly updated according to one's own past claim history. An application to vehicle insurance claims is provided.
    Keywords: Sequential forecasting and pricing,Mixed data,polynomial expansion,random effect,sequential forecast- ing/pricing
    Date: 2019
  9. By: Krüger, Jens; Ruths Sion, Sebastian
    Abstract: In this paper we document the results of a forecast evaluation exercise for the real world price of crude oil using VAR models estimated by sparse (regularization) estimators. These methods have the property to constrain single parameters to zero. We find that estimating VARs with three core variables (real price of oil, index of global real economic activity, change in global crude oil production) by the sparse methods is associated with substantial reductions of forecast errors. The transformation of the variables (taking logs or differences) is also crucial. Extending the VARs by further variables is not associated with additonal gains in forecast performance as is the application of impulse indicator saturation before the estimation.
    Date: 2019–12
  10. By: Francisco Arroyo Marioli; Francisco Bullano; Jorge Fornero; Roberto Zúñiga
    Abstract: The semi-structural gap forecasting (MSEP) model is the new gap model used by the Central Bank of Chile to forecast key macroeconomics variables. This document provides the technical details of this model including equations, estimated parameters and transmission mechanisms. The model has been improved relative to its initial version along several dimensions: (i) The parameters have been estimated with Bayesian methods; (ii) it separates core inflation into tradable and non-tradable inflation, linking each component to fundamental drivers; (iii) it explicitly specifies the empirical relationships between terms of trade and real exchange rate. We found that for a typical monetary policy shocks there are similar effects in comparison with the former MEP model.
    Date: 2020–01
  11. By: Ioannou, Petros; Monteiro, Fernando Valladares
    Abstract: The lack of appropriate and convenient truck parking locations has been identified as a major safety, cost, and environmental issue in both the United States and the European Union. Without access to appropriate parking locations, drivers might be forced to either drive while tired, increasing the risk of accidents, or park illegally in unsafe locations, posing a potential safety hazard to themselves and other drivers. The parking shortage also impacts shipment costs and the environment, since drivers burn more fuel while looking for places to park or idling their engines to provide cab power when parked in inappropriate locations. This research brief summarizes findings from the associated project, the objective of which was to generate parking assist algorithms that can help truck drivers better plan their trips. By providing information about parking availability, the researchers hope to induce truck drivers to better distribute themselves among existing rest areas. This would decrease the peak demand in the most popular truck stops and attenuate the problems caused by the parking shortage. View the NCST Project Webpage
    Keywords: Engineering, Algorithms, Intelligent transportation systems, Mathematical prediction, Parking, Parking facilities, Traffic forecasting, Truck stops, Trucking
    Date: 2020–02–01

This nep-for issue is ©2020 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.