nep-for New Economics Papers
on Forecasting
Issue of 2021‒02‒15
six papers chosen by
Rob J Hyndman
Monash University

  1. Modelling and forecasting inflation rate in Nigeria using ARIMA models By Olalude, Gbenga Adelekan; Olayinka, Hammed Abiola; Ankeli, Uchechi Constance
  2. Designing multi-model applications with surrogate forecast systems By Smith, Leonard A.; Du, Hailiang; Higgins, Sarah
  3. Nowcasting GDP and its Components in a Data-rich Environment: the Merits of the Indirect Approach By Alessandro Giovannelli; Tommaso Proietti; Ambra Citton; Ottavio Ricchi; Cristian Tegami; Cristina Tinti
  4. Forecasting Commodity Markets Volatility: HAR or Rough? By Mesias Alfeus; Christina Sklibosios Nikitopoulos
  5. Nowcasting Monthly GDP with Big Data: a Model Averaging Approach By Tommaso Proietti; Alessandro Giovannelli
  6. Modeling and Forecasting Gold Prices By Quarm, Richmond Sam; Busharads, Mohamed Osman Elamin; Institute of Research, Asian

  1. By: Olalude, Gbenga Adelekan; Olayinka, Hammed Abiola; Ankeli, Uchechi Constance
    Abstract: This study modelled and forecast inflation in Nigeria using the monthly Inflation rate series that spanned January 2003 to October 2020 and provided three years monthly forecast for the inflation rate in Nigeria. We examined 169 ARMA, 169 ARIMA, 1521 SARMA, and 1521 SARIMA models to identify the most appropriate model for modelling the inflation rate in Nigeria. Our findings indicate that out of the 3380 models examined, SARMA (3, 3) x (1, 2)12 is the best model for forecasting the monthly inflation rate in Nigeria. We selected the model based on the lowest Akaike Information Criteria (AIC) and Schwarz Information Criterion (SIC) values, volatility, goodness of fit, and forecast accuracy measures, such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE). The AIC and SIC of the model are 3.3992 and 3.5722, respectively with an adjusted R2 value of 0.916. Our diagnostic tests (Autocorrelation and Normality of Residuals) and forecast accuracy measures indicate that the presented model, SARMA (3, 3)(1, 2)12, is good and reliable for forecasting. Finally, the three years monthly forecast was made, which shows that the Inflation rate in Nigeria would continue to decrease but maintain a 2 digits value for the next two years, but is likely to rise again in 2023. This study is of great relevance to policymakers as it provides a foresight of the likely future inflation rates in Nigeria. Keywords: Inflation; Modelling, Forecasting; ARMA; ARIMA; SARMA; SARIMA;
    Keywords: Inflation; Modelling; Forecasting; ARMA; ARIMA; SARMA; SARIMA
    JEL: C22 C52 C53 E31 E37 E47
    Date: 2020
  2. By: Smith, Leonard A.; Du, Hailiang; Higgins, Sarah
    Abstract: Probabilistic forecasting is common in a wide variety of fields including geoscience, social science, and finance. It is sometimes the case that one has multiple probability forecasts for the same target.How is the information in these multiple nonlinear forecast systems best "combined"? Assuming stationarity, in the limit of a very large forecast-outcome archive, each model-based probability density function can be weighted to form a "multimodel forecast" that will, in expectation, provide at least as much information as the most informative single model forecast system. If one of the forecast systems yields a probability distribution that reflects the distribution from which the outcome will be drawn, Bayesian model averaging will identify this forecast system as the preferred system in the limit as the number of forecast-outcome pairs goes to infinity. In many applications, like those of seasonal weather forecasting, data are precious; the archive is often limited to fewer than 26 entries. In addition, no perfect model is in hand. It is shown that in this case forming a single "multimodel probabilistic forecast" can be expected to provemisleading. These issues are investigated in the surrogatemodel (here a forecast system) regime,where using probabilistic forecasts of a simplemathematical systemallowsmany limiting behaviors of forecast systems to be quantified and compared with those undermore realistic conditions.
    Keywords: EP/K013661/1; EP/K03832X/1)
    JEL: C1
    Date: 2020–06–01
  3. By: Alessandro Giovannelli (University of L'Aquila); Tommaso Proietti (DEF & CEIS, Università di Roma "Tor Vergata"); Ambra Citton (Ministero dell'Economia e delle Finanze); Ottavio Ricchi (Ministero dell'Economia e delle Finanze); Cristian Tegami (Sogei SpA); Cristina Tinti (Ministero dell'Economia e delle Finanze)
    Abstract: The national accounts provide a coherent and exaustive description of the current state of the economy, but are available at the quarterly frequency and are released with a nonignorable publication lag. The paper proposes and illustrates a method for nowcasting and forecasting the sixteen main components of Gross Domestic Product (GDP) by output and expenditure type at the monthly frequency, using a high-dimensional set of monthly economic indicators spanning the space of the common macroeconomic and financial factors. The projection on the common space is carried out by combining the individual nowcasts and forecasts arising from all possible bivariate models of the unobserved monthly GDP component and the observed monthly indicator. We discuss several pooling strategies and we select the one showing the best predictive performance according to a pseudo real time forecasting experiment. Monthly GDP can be indirectly estimated by the contemporaneous aggregation of the value added of the different industries and of the expenditure components. This enables the comparative assessment of the indirect nowcasts and forecasts vis-à-vis the direct approach and a growth accounting exercise. Our approach meets the challenges posed by the dimensionality, since it can handle a large number of time series with a complexity that increases linearly with the cross-sectional dimension, while retaining the essential heterogeneity of the information about the macroeconomy. The application to the Italian case leads to several interesting discoveries concerning the time-varying predictive content of the information carried by the monthly indicators.
    Keywords: Mixed-Frequency Data, Dynamic Factor Models, Growth Accounting, Model Averaging, Ledoit-Wolf Shrinkage.
    JEL: C32 C52 C53 E37
    Date: 2020–05–30
  4. By: Mesias Alfeus; Christina Sklibosios Nikitopoulos (Finance Discipline Group, UTS Business School, University of Technology Sydney)
    Abstract: Commodity is one of the most volatile markets and forecasting its volatility is an issue of paramount importance. We study the dynamics of the commodity markets volatility by employing fractional stochastic volatility and heterogeneous autoregressive (HAR) models. Based on a high-frequency futures price dataset of 22 commodities, we confirm that the volatility of commodity markets is rough and volatility components over different horizons are economically and statistically significant. Long memory with anti-persistence is evident across all commodities, with weekly volatility dominating in most commodity markets and daily volatility for oil and gold markets. HAR models display a clear advantage in forecasting performance compared to fractional volatility models.
    Keywords: commodity markets; realized volatility; fractional Brownian motion; HAR; volatility forecast
    JEL: C20 C53 C58 G13 Q02
    Date: 2020–12–01
  5. By: Tommaso Proietti (CEIS & DEF, University of Rome "Tor Vergata"); Alessandro Giovannelli (DEF, University of Rome "Tor Vergata")
    Abstract: Gross domestic product (GDP) is the most comprehensive and authoritative measure of economic activity. The macroeconomic literature has focused on nowcasting and forecasting this measure at the monthly frequency, using related high frequency indicators. We address the issue of estimating monthly gross domestic product using a large dimensional set of monthly indicators, by pooling the disaggregate estimates arising from simple and feasible bivariate models that consider one indicator at a time in conjunction to GDP. Our base model handles mixed frequency data and ragged-edge data structure with any pattern of missingness. Our methodology enables to distill the common component of the available economic indicators, so that the monthly GDP estimates arise from the projection of the quarterly figures on the space spanned by the common component. The weights used for the combination reflect the ability to nowcast quarterly GDP and are obtained as a function of the regularized estimator of the high-dimensional covariance matrix of the nowcasting errors. A recursive nowcasting and forecasting experiment illustrates that the optimal weights adapt to the information set available in real time and vary according to the phase of the business cycle.
    Keywords: Mixed-Frequency Data, Dynamic Factor Models, State Space Models,Shrinkage
    JEL: C32 C52 C53 E37
    Date: 2020–05–12
  6. By: Quarm, Richmond Sam; Busharads, Mohamed Osman Elamin; Institute of Research, Asian
    Abstract: The aim of this paper is to explore the reasons of gold price volatility. It analyses the information function of the gold future market by open interest contracts as speculation effect, and further fundamental factors including inflation, Chinese yuan per dollar, Japanese yen per dollar, dollar per euro, interest rate, oil price, and stock price, in the short-run. The study proceeds to build a Dynamic OLS model for long-run equilibrium to produce reliable gold price forecasts using the following variables: gold demand, gold supply, inflation, USD/SDR exchange rate, speculation, interest rate, oil price, and stock prices. Findings prove that in the short-run, changes in gold price does granger cause changes in open interest, and changes in Japanese yen per dollar does granger cause changes in gold price. However, in the long-run, the results prove that gold demand, gold supply, USD/SDR exchange rate, inflation, speculation, interest rate, and oil price are associated in a long-run relationship.
    Date: 2020–12–26

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.