nep-for New Economics Papers
on Forecasting
Issue of 2018‒01‒15
six papers chosen by
Rob J Hyndman
Monash University

  1. Working Paper – WP/16/01- Nowcasting Real GDP growth in South Africa By Alain Kabundi; Elmarie Nel; Franz Ruch
  2. Financial variables and macroeconomic forecast errors By Barnes, Michelle L.; Olivei, Giovanni P.
  3. Working Paper – WP/16/08- Modeling and Forecasting Daily Financial and Commodity Term Structures- A Unified Global Approach By Shakill Hassan; Leonardo Morales-Arias
  4. Forecasting Base Metal Prices with Commodity Currencies By Pincheira, Pablo; Hardy, Nicolas
  5. Residential investment and recession predictability By Knut Are Aastveit; Author-Name: André K. Anundsen; Eyo I. Herstad
  6. Deep Learning for Forecasting Stock Returns in the Cross-Section By Masaya Abe; Hideki Nakayama

  1. By: Alain Kabundi; Elmarie Nel; Franz Ruch
    Abstract: This paper uses nowcasting to forecast real GDP growth in South Africa from 2010Q1 to 2014Q3 in real time. Such an approach exploits the flow of high-frequency information underlying the state of the economy. It overcomes one of the major challenges faced by forecasters, policymakers, and economic agents - having a clear view of the state of the economy in real time. This is often not the case as many economic variables are only available at low frequency and with considerable lags, making it difficult to have information on the state of the economy even after the end of the quarter. The pseudo out-of-sample forecasts show that the nowcasting model’s performance is comparable to those of professional forecasters even though the latter enhance their forecasting accuracy with judgement. The nowcast model also outperforms all other benchmark models by a significant margin.
    Date: 2016–02–03
  2. By: Barnes, Michelle L. (Federal Reserve Bank of Boston); Olivei, Giovanni P. (Federal Reserve Bank of Boston)
    Abstract: A large set of financial variables has only limited power to predict a latent factor common to the year-ahead forecast errors for real Gross Domestic Product (GDP) growth, the unemployment rate, and Consumer Price Index (CPI) inflation for three sets of professional forecasters: the Federal Reserve’s Greenbook, the Survey of Professional Forecasters (SPF), and the Blue Chip Consensus Forecasts. Even when a financial variable appears to be fairly robust across sample periods in explaining the latent factor, from an economic standpoint its contribution appears modest. Still, several financial variables retain economic significance over certain subsamples; when non-linear effects are accounted for, these variables have an improved ability to consistently predict the latent factor over the business cycle.
    Keywords: forecast errors; macroeconomy; financial variables; threshold estimation; business cycle
    JEL: C24 C53 E37 E44 E50 G01 G17
    Date: 2017–10–31
  3. By: Shakill Hassan; Leonardo Morales-Arias
    Abstract: In this article we propose a dynamic factor framework for modeling and forecasting financial and commodity term structures in a unified global setting. The novelty of our approach is that it exploits a large set of information (i.e. data properties, time and forward dimensions, and cross-country, market, sector and weather dimensions) summarized in a set of heteroskedastic components that have a clear time series interpretation and that can be modeled dynamically to generate forecasts in real-time. The approach is motivated by evidence of rising financial integration, and interdependence between commodity and asset markets. We employ a battery of in-sample and out-of-sample techniques to evaluate our framework and concentrate on relevant statistical and economic performance measures. To preview our results with practical implications, we find that our framework provides significant in-sample information in terms of product specific factors and commonalities driving commodity and financial markets. Moreover, the specification proposed for modelling the dynamics of financial and commodity term structures generates accurate out-of-sample interval and point forecasts and leads to variance reduction when hedging a portfolio made up of spot and futures contracts.
    Date: 2016–07–11
  4. By: Pincheira, Pablo; Hardy, Nicolas
    Abstract: In this paper we show that the Chilean exchange rate has the ability to predict the returns of the London Metal Exchange Index and of the six primary non-ferrous metals that are part of the index: aluminum, copper, lead, nickel, tin and zinc. The economic relationship hinges on the present-value theory for exchange rates, a floating exchange rate regime and the fact that copper represents about a half of Chilean exports and nearly 45% of Foreign Direct Investment. Consequently, the Chilean peso is heavily affected by fluctuations in the copper price. As all six base metal prices show an important comovement, we test whether the relationship between copper prices and Chilean exchange rates also holds true when it comes to the six primary non-ferrous metals. We find interesting evidence of predictability both in-sample and out-of-sample. Our paper is part of a growing literature that in the recent years has evaluated and called into question the ability of commodity currencies to forecast commodity prices.
    Keywords: Forecasting, commodities prices, univariate time-series models, out-of-sample comparison, exchange rates, copper, primary non-ferrous metals.
    JEL: E3 E31 E32 E37 E5 E52 F0 F00 F01 F21 F31 F47 G1 G12 G14 G15 G17 Q3 Q32 Q4 Q41 Q47
    Date: 2018–01–02
  5. By: Knut Are Aastveit; Author-Name: André K. Anundsen; Eyo I. Herstad
    Abstract: We assess the importance of residential investment in predicting economic recessions for an unbalanced panel of 12 OECD countries over the period 1960Q1–2014Q4. Our approach is to estimate various probit models with di?erent leading indicators and evaluate their relative prediction accuracy using the receiver operating characteristic curve. We document that residential investment contains information useful in predicting recessions both in-sample and out-of-sample. This result is robust to adding typical leading indicators, such as the term spread, stock prices, consumer confidence surveys and oil prices. It is shown that residential investment is particularly useful in predicting recessions for countries with high home-ownership rates. Finally, in a separate exercise for the US economy, we show that the predictive ability of residential investment is robust to employing real-time data.
    Keywords: Recession predictability, Housing, Leading indicators, Real-time data
    Date: 2017–12
  6. By: Masaya Abe; Hideki Nakayama
    Abstract: Many studies have been undertaken by using machine learning techniques, including neural networks, to predict stock returns. Recently, a method known as deep learning, which achieves high performance mainly in image recognition and speech recognition, has attracted attention in the machine learning field. This paper implements deep learning to predict one-month-ahead stock returns in the cross-section in the Japanese stock market and investigates the performance of the method. Our results show that deep neural networks generally outperform shallow neural networks, and the best networks also outperform representative machine learning models. These results indicate that deep learning shows promise as a skillful machine learning method to predict stock returns in the cross-section.
    Date: 2018–01

This nep-for issue is ©2018 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.