nep-for New Economics Papers
on Forecasting
Issue of 2021‒05‒31
twelve papers chosen by
Rob J Hyndman
Monash University

  1. Predicting returns and dividend growth - the role of non-Gaussian innovations By Kiss, Tamás; Mazur, Stepan; Nguyen, Hoang
  2. Comparing Forecast Performance with State Dependence By Odendahl, Florens; Rossi, Barbara; Sekhposyan, Tatevik
  3. Modeling and Forecasting Macroeconomic Downside Risk By De Polis, Andrea; Delle Monache, Davide; Petrella, Ivan
  4. Forecasting Base Metal Prices with an International Stock Index By Pincheira, Pablo; Hardy, Nicolas; Bentancor, Andrea; Henriquez, Cristóbal; Tapia, Ignacio
  5. Forecasting with Shadow-Rate VARs By Andrea Carriero; Todd E. Clark; Massimiliano Marcellino; Elmar Mertens
  6. Forecasting with fractional Brownian motion: a financial perspective By Matthieu Garcin
  7. Forecasting the Covid-19 recession and recovery: Lessons from the financial crisis By Foroni, Claudia; Marcellino, Massimiliano; Stevanovic, Dalibor
  8. Forecasting Output Growth of Advanced Economies Over Eight Centuries: The Role of Gold Market Volatility as a Proxy of Global Uncertainty By Afees A. Salisu; Rangan Gupta; Sayar Karmakar; Sonali Das
  9. Multi-Horizon Forecasting for Limit Order Books: Novel Deep Learning Approaches and Hardware Acceleration using Intelligent Processing Units By Zihao Zhang; Stefan Zohren
  10. Financial Time Series Analysis and Forecasting with HHT Feature Generation and Machine Learning By Tim Leung; Theodore Zhao
  11. Forecasting recovery from COVID-19 using financial data: An application to Viet Nam By Jesse Lastunen; Matteo Richiardi
  12. GDP Modelling and Forecasting Using ARIMA. An Empirical Assessment for Innovative Economy Formation By VINTU, Denis

  1. By: Kiss, Tamás (Örebro University School of Business); Mazur, Stepan (Örebro University School of Business); Nguyen, Hoang (Örebro University School of Business)
    Abstract: In this paper we assess whether exible modelling of innovations impact the predictive performance of the dividend price ratio for returns and dividend growth. Using Bayesian vector autoregressions we allow for stochastic volatility, heavy tails and skewness in the innovations. Our results suggest that point forecasts are barely affected by these features, suggesting that workhorse models on predictability are sufficient. For density forecasts, however, we finnd that stochastic volatility substantially improves the forecasting performance.
    Keywords: Bayesian VAR; Dividend Growth Predictability; Predictive Regression; Return Predictability
    JEL: C11 C58 G12
    Date: 2021–05–24
  2. By: Odendahl, Florens; Rossi, Barbara; Sekhposyan, Tatevik
    Abstract: We propose a novel forecast comparison methodology to evaluate models' relative forecasting performance when the latter is a state-dependent function of economic variables. In our bench¬mark case, the relative forecasting performance, measured by the forecast loss differential, is modeled via a threshold model. Importantly, we allow the threshold that triggers the switch from one state to the next to be unknown, leading to a non-standard test statistic due to the presence of a nuisance parameter. Existing tests either assume a constant out-of-sample forecast performance or use non-parametric techniques robust to time-variation; consequently, they may lack power against state-dependent predictability. Importantly, our approach is applicable to point forecasts as well as predictive densities. Monte Carlo results suggest that our proposed test statistics perform well in ï¬ nite samples and have better power than existing tests in selecting the best forecasting model in the presence of state dependence. Our test statistics uncover "pockets of predictability" in U.S. equity premia forecasts; the pockets are a state-dependent function of stock market volatility. Models using economic predictors perform signiï¬ cantly worse than a simple mean forecast in periods of high volatility, but, in periods of low volatility, the use of economic predictors may lead to small forecast improvements.
    Keywords: Forecast evaluation; Pockets of Predictability; State Dependence
    JEL: C52 C53 G17
    Date: 2020–08
  3. By: De Polis, Andrea; Delle Monache, Davide; Petrella, Ivan
    Abstract: We document a substantial increase in downside risk to US economic growth over the last 30 years. By modeling secular trends and cyclical changes of the predictive density of GDP growth, we recover an accelerating decline in the skewness of the conditional distributions, with significant, procyclical variations. Decreasing trend-skewness, turning negative in the aftermath of the Great Recession, is associated with the long-run growth slowdown stared in the early 2000s. Short-run skewness fluctuation imply negatively skewed predictive densities ahead, and during recessions, often anticipated by deteriorating financial conditions, while positively skewed distributions characterize expansions. The model delivers competitive out-of-sample (point, density and tail) forecasts, improving upon standard benchmarks, due to financial conditions providing strong signals of increasing downside risk.
    Keywords: Business cycle; Downside risk; financial conditions; score driven models; Skewness
    JEL: C53 E32 E44
    Date: 2020–07
  4. By: Pincheira, Pablo; Hardy, Nicolas; Bentancor, Andrea; Henriquez, Cristóbal; Tapia, Ignacio
    Abstract: In this paper we show that the MSCI ACWI Metals and Mining Index has the ability to predict base metal prices. We use both in-sample and out-of-sample exercises to conduct such examination. The theoretical underpinning of these results relies on the present-value model for stock-price determination. This model has the implication of Granger causality from stock prices to their key determinants. In the case of metal and mining producers, one of the key elements determining the value of these firms is the price of the commodity they produce and export. Our results are consistent with this theoretical framework, as forecasts based on a model including the MSCI index outperform, in terms of Mean Squared Prediction Error, forecasts that do not use the information contained in that index.
    Keywords: Forecasting, commodities, base metals, univariate time-series models, out-of-sample comparison, base metal equity securities.
    JEL: C10 C12 C2 C22 C4 C5 C52 C53 C58 C6 E0 E3 E31 E32 E37 E4 E47 E6 F3 F31 F4 F44 F47 G0 G1 G12 G17
    Date: 2021–05–18
  5. By: Andrea Carriero; Todd E. Clark; Massimiliano Marcellino; Elmar Mertens
    Abstract: Interest rate data are an important element of macroeconomic forecasting. Projections of future interest rates are not only an important product themselves, but also typically matter for forecasting other macroeconomic and financial variables. A popular class of forecasting models is linear vector autoregressions (VARs) that include shorter- and longer-term interest rates. However, in a number of economies, at least shorter-term interest rates have now been stuck for years at or near their effective lower bound (ELB), with longer-term rates drifting toward the constraint as well. In such an environment, linear forecasting models that ignore the ELB constraint on nominal interest rates appear inept. To handle the ELB on interest rates, we model observed rates as censored observations of a latent shadow-rate process in an otherwise standard VAR setup. The shadow rates are assumed to be equal to observed rates when above the ELB. Point and density forecasts for interest rates (short term and long term) constructed from a shadow-rate VAR for the US since 2009 are superior to predictions from a standard VAR that ignores the ELB. For other indicators of financial conditions and measures of economic activity and inflation, the accuracy of forecasts from our shadow-rate specification is on par with a standard VAR that ignores the ELB.
    Keywords: Macroeconomic forecasting; effective lower bound; term structure; censored observations
    JEL: C34 C53 E17 E37 E43 E47
    Date: 2021–03–29
  6. By: Matthieu Garcin (Research Center - Léonard de Vinci Pôle Universitaire - De Vinci Research Center)
    Abstract: The fractional Brownian motion (fBm) extends the standard Brownian motion by introducing some dependence between non-overlapping increments. Consequently, if one considers for example that log-prices follow an fBm, one can exploit the non-Markovian nature of the fBm to forecast future states of the process and make statistical arbitrages. We provide new insights into forecasting an fBm, by proposing theoretical formulas for accuracy metrics relevant to a systematic trader, from the hit ratio to the expected gain and risk of a simple strategy. In addition, we answer some key questions about optimizing trading strategies in the fBm framework: Which lagged increments of the fBm, observed in discrete time, are to be considered? If the predicted increment is close to zero, up to which threshold is it more profitable not to invest? We also propose empirical applications on high-frequency FX rates, as well as on realized volatility series, exploring the rough volatility concept in a forecasting perspective.
    Keywords: rough volatility,foreign-exchange rates,fractional Brownian motion,Hurst exponent,systematic trading
    Date: 2021–05–19
  7. By: Foroni, Claudia; Marcellino, Massimiliano; Stevanovic, Dalibor
    Abstract: We consider simple methods to improve the growth nowcasts and forecasts obtained by mixed frequency MIDAS and UMIDAS models with a variety of indicators during the Covid-19 crisis and recovery period, such as combining forecasts across various specifications for the same model and/or across different models, extending the model specification by adding MA terms, enhancing the estimation method by taking a similarity approach, and adjusting the forecasts to put them back on track by a specific form of intercept correction. Among all these methods, adjusting the original nowcasts and forecasts by an amount similar to the nowcast and forecast errors made during the financial crisis and following recovery seems to produce the best results for the US, notwithstanding the different source and characteristics of the financial crisis. In particular, the adjusted growth nowcasts for 2020Q1 get closer to the actual value, and the adjusted forecasts based on alternative indicators become much more similar, all unfortunately indicating a much slower recovery than without adjustment and very persistent negative effects on trend growth. Similar findings emerge also for the other G7 countries.
    Date: 2020–07
  8. By: Afees A. Salisu (Centre for Econometric & Allied Research, University of Ibadan, Ibadan, Nigeria); Rangan Gupta (Department of Economics, University of Pretoria, Private Bag X20, Hatfield, 0028, South Africa); Sayar Karmakar (Department of Statistics, University of Florida, 230 Newell Drive, Gainesville, FL 32601, USA); Sonali Das (Department of Business Management, University of Pretoria, Private Bag X20, Hatfield 0028, South Africa)
    Abstract: In this paper we develop a proxy for global uncertainty based on the volatility of gold market over the annual period of 1311 to 2019, and then use this proxy metric to forecast historical growthrates for eight advance economies namely, France, Germany, Holland, Italy, Japan, Spain, the United Kingdom (UK), and the United States (US). We find that for the within-sample period, uncertainty negatively impacts output growth, but more importantly, over the out-of-sample period, gold market volatility produces statistically significant forecasting gains. Our findings are robust to an alternative measure of uncertainty based on the volatility of the changes in long-term sovereign real-rates over 1315 to 2019. These historical results have important implications for investors and policymakers in the current context in which high frequency gold price data is available.
    Keywords: Historical output growth, advance economies, gold market volatility, forecasting
    JEL: C22 C53 E32 Q02
    Date: 2021–05
  9. By: Zihao Zhang; Stefan Zohren
    Abstract: We design multi-horizon forecasting models for limit order book (LOB) data by using deep learning techniques. Unlike standard structures where a single prediction is made, we adopt encoder-decoder models with sequence-to-sequence and Attention mechanisms, to generate a forecasting path. Our methods achieve comparable performance to state-of-art algorithms at short prediction horizons. Importantly, they outperform when generating predictions over long horizons by leveraging the multi-horizon setup. Given that encoder-decoder models rely on recurrent neural layers, they generally suffer from a slow training process. To remedy this, we experiment with utilising novel hardware, so-called Intelligent Processing Units (IPUs) produced by Graphcore. IPUs are specifically designed for machine intelligence workload with the aim to speed up the computation process. We show that in our setup this leads to significantly faster training times when compared to training models with GPUs.
    Date: 2021–05
  10. By: Tim Leung; Theodore Zhao
    Abstract: We present the method of complementary ensemble empirical mode decomposition (CEEMD) and Hilbert-Huang transform (HHT) for analyzing nonstationary financial time series. This noise-assisted approach decomposes any time series into a number of intrinsic mode functions, along with the corresponding instantaneous amplitudes and instantaneous frequencies. Different combinations of modes allow us to reconstruct the time series using components of different timescales. We then apply Hilbert spectral analysis to define and compute the associated instantaneous energy-frequency spectrum to illustrate the properties of various timescales embedded in the original time series. Using HHT, we generate a collection of new features and integrate them into machine learning models, such as regression tree ensemble, support vector machine (SVM), and long short-term memory (LSTM) neural network. Using empirical financial data, we compare several HHT-enhanced machine learning models in terms of forecasting performance.
    Date: 2021–05
  11. By: Jesse Lastunen; Matteo Richiardi
    Abstract: We develop a new methodology to nowcast the effects of the COVID-19 crisis and forecast its evolution in small, export-oriented countries. To this aim, we exploit variation in financial indexes at the industry level and relate them to the expected duration of the crisis for each industry, under the assumption that the main shocks to financial prices in recent months have come from COVID-19.
    Keywords: COVID-19, Pandemic, Forecasting, Finance, Viet Nam, Shocks
    Date: 2021
  12. By: VINTU, Denis
    Abstract: This article reconsiders the developing of a new forecast model using the interrupted timeseries of the gross domestic product for the Republic of Moldova. The theme arises from a first need to redefine, economic growth in the context of increasing globalization but also the complexity of commercial transactions. The forecasting method used is based on ARIMA each model partly emphasizing the urgent need to redefine, the economic growth in the context of the Association Agreement (AA) with the EU, which includes a Comprehensive Free Trade Agreement (2014) but also future prospects of integration among the countries with an average degree of development. The technique used comes to bring novelty in the field of forecasting, as an alternative to the one which should be —, a simultaneous equations method and traditional VAR. The policy and practical implications of the results are the strengths. The limits are due to the high degree of risk and uncertainty, which is due to the low degree of real convergence of the economy, but also to other factors such as the regional context, the lack of openness of the economy, the diversification of exports and services. The degree of complexity arises from the adaptation and study of the chronological interrupted series 1967−2019 for the branch – information and communications, subgroup GDP, categories of resources, which themselves have specific asymmetries and nuances. The basic ARIMA equations are generally used in conjunction with three sets of assumptions regarding the formation of the gross domestic product, referring to the elasticity of aggregate demand or excess sensitivity supply in the goods and labour markets. Another hypothesis concerns the rigid wage and sticky prices, including deflation with an positive output gap only in the telecom market. Also, the salary is rigid, while the price level is adjusted based on the market of goods and commodities, so that the excess supply appears only in the labour market. Finally, in a third assumption, both markets are assumed to be mutually adjusted. The multipliers of fiscal and monetary policy, besides the conclusions that can be drawn about economic policy, are obviously different in these three assumptions. The article presents a synthetic model that supports the three particular sub-regimes of assumptions of a single adapted ARIMA model, namely the trajectory of New Keynesian Small and Closed Economy Model – a balance in the goods and services, the labour market and the national financial system. In conclusion, the model aims not only to redefine the area of macroeconomic forecasting but also to offer a future perspective of adopting combined techniques such as the Stochastic Dynamic General Equilibrium (K-SDGE) Model with sticky prices and wages – technique, but also the scenario method. This framework is appealing because it has straight forward model setup, transparent mechanisms, sharp empirical analysis, and multiple important applications such as rational expectations.
    Keywords: economic growth and aggregate productivity, the gross domestic product, innovation and communications, cross-country output convergence, prediction and forecasting methods,time series analysis and modelling, ARIMA modelling, Box-Jenkins method.
    JEL: C12 C14 C22 C53 D62 D84 F15 F21 F61 O10 O30
    Date: 2021–04–23

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.