nep-for New Economics Papers
on Forecasting
Issue of 2015‒09‒26
thirteen papers chosen by
Rob J Hyndman
Monash University

  1. Selection of an estimation window in the presence of data revisions and recent structural breaks By Hännikäinen, Jari
  2. Modelling and Forecasting of Tunisian Current Account: Aggregate versus Disaggregate Approach By Kamel Jlassi
  3. Evaluating Performance of Inflation Forecasting Models of Pakistan By Hanif, Muhammad Nadim; Malik, Muhammad Jahanzeb
  4. The Out-of-Sample Forecasting Performance of Non-Linear Models of Regional Housing Prices in the US By Mehmet Balcilar; Rangan Gupta; Stephen M. Miller
  5. Dealing with Financial Instability under a DSGE modeling approach with Banking Intermediation: a forecastability analysis versus TVP-VARs By Bekiros, Stelios D.; Cardani, Roberta; Paccagnini, Alessia; Villa, Stefania
  6. A large-scale application of Stata's forecast suite: challenges and potential By Christopher F Baum
  7. Predictable Recoveries By Cai, Xiaoming; Den Haan, Wouter; Pinder, Jonathan
  8. An Alternative Reference Scenario for Global CO2Emissions from Fuel Consumption: An ARFIMA Approach By José M. Belbute; Alfredo Marvão Pereira
  9. (Mis-)Predicted Subjective Well-Being Following Life Events By Reto Odermatt; Alois Stutzer
  10. Comparing predictive accuracy in small samples By Laura Coroneo; Fabrizio Iacone
  11. Determinants of Potato Prices and its Forecasting: A Case Study of Punjab, Pakistan By Anwar, Dr. Mumtaz; Shabbir, Dr. Ghulam; Shahid, M. Hassam; Samreen, Wajiha
  12. A Comparative Study of Stock Price Forecasting using nonlinear models By Lawrence Xaba; Ntebogang Moroke; Johnson Arkaah; Charlemagne Pooe
  13. Forecasting South African Gold Sales: The Box-Jenkins Methodology By Johannes Tshepiso Tsoku; Nonofo Phokontsi; Daniel Metsileng

  1. By: Hännikäinen, Jari
    Abstract: In this paper, we analyze the forecasting performance of a set of widely used window selection methods in the presence of data revisions and recent structural breaks. Our Monte Carlo and empirical results for U.S. real GDP and inflation show that the expanding window estimator often yields the most accurate forecasts after a recent break. It performs well regardless of whether the revisions are news or noise, or whether we forecast first-release or final values. We find that the differences in the forecasting accuracy are large in practice, especially when we forecast inflation after the break of the early 1980s.
    Keywords: Recent structural break, choice of estimation window, forecasting, real-time data
    JEL: C22 C53 C82
    Date: 2015–09–18
  2. By: Kamel Jlassi (Central Bank of Tunisia)
    Abstract: While there is considerable literature attempting to model current account, there is fewer studies to forecast current account balance. This study gives a comprehensive way to model and predict current account deficit (CAD) by evaluating the forecasting performance of direct and indirect approach. At disaggregated level, I use two variants to model current account component; in the first alternative I apply different ARIMA model with exogenous variables (ARIMA-X) to account for the pattern of the data and exogenous factors. In the second alternative, I integrate the cointegration relationship between exports and imports with ARIMA-X models. With respect to the direct approach, I use error correction model to allow for dynamics in current account. The data used spans from January 2000 to December 2014 and comes from Central Bank of Tunisia, Tunisian National Institute of Statistic, and OECD database. I find that for one-step ahead forecast ARIMA-X and reduced form model produce accurate forecast but with respect to dynamic forecast, direct method is more accurate comparatively to ARIMA-X. When cointegrating relationship between exports and imports is combined with ARIMA-X models, indirect approach outperforms direct approach. I also show that, as volatility of underlying components increase disaggregate approach using time series models become less reliable. In addition, I found that current account is mainly affected by local GDP, trade openness, fiscal deficit, exchange rate, credit to private sector and partner GDP. Estimation of ECM indicates that persistent effect is high and can take more than three quarters to die out. In addition I assess the performance of direct and indirect approach over time using naïve approach as benchmark. It appears that the MSE of naïve approach lies between direct and indirect approach in average up to horizon 12, but then worsen.
    Keywords: Aggregate and Disaggregate Approach, Cointegration, Error Correction Model, Time Series Models, Current Account Forecast, One-step ahead and Dynamic Forecast.
    Date: 2015–06–12
  3. By: Hanif, Muhammad Nadim; Malik, Muhammad Jahanzeb
    Abstract: This study compares the forecasting performance of various models of inflation for a developing country estimated over the period of last two decades. Performance is measured at different forecast horizons (up to 24 months ahead) and for different time periods when inflation is low, high and moderate (in the context of Pakistan economy). Performance is considered relative to the best amongst the three usually used forecast evaluation benchmarks – random walk, ARIMA and AR(1) models. We find forecasts from ARDL modeling and certain combinations of point forecasts better than the best benchmark model, the random walk model, as well as structural VAR and Bayesian VAR models for forecasting inflation for Pakistan. For low inflation regime, upper trimmed average of the point forecasts out performs any model based forecasting for short period of time. For longer period, use of an ARDL model is the best choice. For moderate inflation regime different ways to average various models’ point forecasts turn out to be the best for all inflation forecasting horizons. The most important case of high inflation regime was best forecasted by ARDL approach for all the periods up to 24 months ahead. In overall, we can say that forecasting performance of different approaches is state dependent for the case of developing countries, like Pakistan, where inflation is occasionally high and volatile.
    Keywords: Inflation, Forecast Evaluation, Random Walk model, AR(1) model, ARIMA model, ARDL model, Structural VAR model, Bayesian VAR model, Trimmed Average
    JEL: C52 E31 E37
    Date: 2015–09–22
  4. By: Mehmet Balcilar (Department of Economics, Eastern Mediterranean University); Rangan Gupta (Department of Economics, University of Pretoria); Stephen M. Miller (Department of Economics, University of Nevada and University of Connecticut)
    Abstract: This paper provides out-of-sample forecasts of linear and non-linear models of US and Census regions housing prices. The forecasts include the traditional point forecasts, but also include interval and density forecasts of the housing price distributions. The non-linear smooth-transition autoregressive model outperforms the linear autoregressive model in point forecasts at longer horizons, but the linear autoregressive model dominates the non-linear smooth-transition autoregressive model at short horizons. In addition, we generally do not find major differences in performance for the interval and density forecasts between the linear and non-linear models. Finally, in a dynamic 25-step ex-ante and interval forecasting design, we, once again, do not find major differences between the linear and nonlinear models.
    Keywords: Forecasting, Linear and non-linear models, US and Census housing price indexes
    JEL: C32 R31
  5. By: Bekiros, Stelios D.; Cardani, Roberta; Paccagnini, Alessia; Villa, Stefania
    Abstract: Recently there has been an increasing awareness on the role that the banking sector can play in macroeconomic activity, especially within the context of the DSGE literature. In this work, we present a DSGE model with financial intermediation as in Gertler and Karadi (2011). The estimation of the shocks and of the structural parameters shows that time-variation can be crucial in the empirical analysis. As DSGE modeling fails to take into account inherent nonlinearities of the economy, we introduce a novel time-varying coefficient state-space estimation method for VAR processes, for homoskedastic and heteroskedastic error structures (TVP-VAR). We conduct an extensive empirical exercise to compare the out-of-sample forecastability of the DSGE model versus standard ARs, VARs, Bayesian VARs and TVP-VARs. We find that the TVP-VAR provides the best forecasting performance for the series of GDP and net worth of financial intermediaries for all steps-ahead, while the DSGE model with the incorporation of a banking sector outperforms the other specifications in forecasting inflation and the federal funds rate at shorter horizons.
    Keywords: DSGE, Extended Kalman Filter, Financial Frictions, Banking Sector
    JEL: C11 C13 E37
    Date: 2015
  6. By: Christopher F Baum (Boston College; DIW Berlin)
    Abstract: Stata 13 added a very important feature for macroeconomists: the forecast suite of commands that implements the definition of a model, consisting of a number of estimated equations and potentially nonlinear identities. Stata’s features include model solution, dynamic forecasting, scenario analysis and stochastic simulation. I report on my attempt to apply the forecast suite to a well-known large-scale macroeconomic model. I discuss the challenges related to use of these features in a much more complex context than that illustrated in the manual’s examples. I will also suggest a number of enhancements that would improve forecast’s capabilities in comparison to other popular forecasting tools.
    Date: 2015–09–16
  7. By: Cai, Xiaoming; Den Haan, Wouter; Pinder, Jonathan
    Abstract: Should an unexpected change in real GNP of x% lead to an x% change in the forecasts of future GNP? The answer could be no even if GNP is a random walk. We show that US economic downturns often go together with predictable short-term recoveries and with changes in long-term GNP forecasts that are substantially smaller than the initial drop. But not always! Essential for our results is that GNP forecasts are not based on a univariate time series model, which is not uncommon. Our alternative forecasts are based on a simple multivariate representation of GNP’s expenditure components.
    Keywords: business cycles; forecasting; unit root
    JEL: C53 E32 E37
    Date: 2015–09
  8. By: José M. Belbute (Department of Economics, University of Évora, Portugal); Alfredo Marvão Pereira (Department of Economics, The College of William and Mary)
    Abstract: We provide alternative reference forecasts for global CO2 emissions based on an ARFIMA model estimated with annual data from 1750 to 2013. These forecasts are free from additional assumptions on demographic and economic variables that are commonly used in reference forecasts, as they only rely on the properties of the underlying stochastic process for CO2emissions, as well as on all the observed information it incorporates. In this sense, these forecasts are more based on fundamentals. Our reference forecast suggests that in 2030, 2040 and 2050, in the absence of any structural changes of any type, CO2 would likely be at about 25%, 34% and 39.9% above 2010 emission levels, respectively. These values are clearly below the levels proposed by other reference scenarios available in the literature. This is important, as it suggests that the ongoing policy goals are actually within much closer reach than what is implied by the standard CO2reference emission scenarios. Having lower and more realistic reference emissions projections not only gives a truer assessment of the policy efforts that are needed, but also highlights the lower costs involved in mitigation efforts, thereby maximizing the likelihood of more widespread energy and environmental policy efforts.
    Keywords: Forecasting, reference scenario, CO2 emissions, long memory, ARFIMA
    JEL: C22 C53 O13 Q47 Q54
    Date: 2015–08–31
  9. By: Reto Odermatt; Alois Stutzer
    Abstract: The correct prediction of how alternative states of the world affect our lives is a cornerstone of economics. We study how accurate people are in predicting their future well-being when facing major life events. Based on individual panel data, we compare people's forecast of their life satisfaction in five years' time to their actual realisations later on. This is done after the individuals experience widowhood, marriage, unemployment or disability. We find systematic prediction errors that are at least partly driven by unforeseen adaptation.
    Keywords: Adaptation, life satisfaction, life events, projection-bias, subjective well-being, utility prediction, unemployment
    Date: 2015
  10. By: Laura Coroneo; Fabrizio Iacone
    Abstract: We consider fixed-b and fixed-m asymptotics for the Diebold and Mariano (1995) test of predictive accuracy. We show that this approach allows to obtain predictive accuracy tests that are correctly sized even in small samples. We apply the alternative asympotics for the Diebold and Mariano (1995) test to evaluate the predictive accuracy of the Survey of Professional Forecasters (SPF) against a simple random walk. Our results show that the predictive ability of the SPF was partially spurious, especially in the last decade.
    Keywords: Diebold and Mariano test, long run variance estimation, fixed-b and fixed-m asymptotic theory, SPF.
    JEL: C12 C32 C53 E17
    Date: 2015–09
  11. By: Anwar, Dr. Mumtaz; Shabbir, Dr. Ghulam; Shahid, M. Hassam; Samreen, Wajiha
    Abstract: Potato figures among the principal crop in Pakistan. This paper describes the determinants of potato prices in Punjab, Pakistan. Annual data for the period 1998-2014 were analyzed to identify factors affecting the prices of potato. Results indicated that temperature and world oil prices were significantly affecting price. Seasonal variation of prices are also analyzed in this paper. This paper also use ARIMA and ARMA model to forecast the prices. These results suggest that temperature increase above the limit will lead to increase in prices and support prices also.
    Keywords: Whole sale potato prices, pricing factors, Seasonal Variation, determinants of potato, ARIMA & ARMA, forecasting.
    JEL: E3 E31 E37
    Date: 2015–09–16
  12. By: Lawrence Xaba (North West University); Ntebogang Moroke (North West University); Johnson Arkaah (North West University); Charlemagne Pooe (South African Reserve Bank)
    Abstract: This study compared the in-sample forecasting accuracy of three forecasting nonlinear models namely: the Smooth Transition Regression (STR) model, the Threshold Autoregressive (TAR) model and the Markov-switching Autoregressive (MS-AR) model. Data used was daily close stock prices of five banks in the South African banking sector and was obtained from the Johannesburg Stock Exchange (JSE). It covered the period from 2010 to 2012 with a total of 563 observations. Nonlinearity and nonstationarity tests used confirmed the validity of the assumptions of the study. The study used model selection criteria, SBC to select the optimal lag order and for the selection of appropriate models. The Mean Square Error (MSE), Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) served as the error measures in evaluating the forecasting ability of the models. The MS-AR models proved to perform well with lower error measures as compared to LSTR and TAR models in most cases.
    Keywords: Stock price, nonlinear time series models, error metrics
    JEL: C10 C32 E32
  13. By: Johannes Tshepiso Tsoku (North West University); Nonofo Phokontsi (North West University); Daniel Metsileng (Department of Health)
    Abstract: The study deals with Box-Jenkins Methodology to forecast South African gold sales. For a resource economy like South Africa where metals and minerals account for a high proportion of GDP and export earnings, the decline in gold sales is very disturbing. Box-Jenkins time series technique was used to perform time series analysis of monthly gold sales for the period January 2000 to June 2013 with the following steps: model identification, model estimation, diagnostic checking and forecasting. Furthermore, the prediction accuracy is tested using mean absolute percentage error (MAPE). From the analysis, a seasonal ARIMA(4,1,4)×(0,1,1)12 was found to be the “best fit model” with an MAPE value of 11% indicating that the model is fit to be used to predict or forecast future gold sales for South Africa. In addition, the forecast values show that there will be a decrease in the overall gold sales for the first six months of 2014. It is hoped that the study will help the public and private sectors to understand the gold sales or output scenario and later plan the gold mining activities in South Africa. Furthermore, it is hoped that this research paper has demonstrated the significance of Box-Jenkins technique for this area of research and that they will be applied in the future.
    Keywords: Gold sales, ARIMA, Box-Jenkins, GDP, MAPE
    JEL: C53 C19 C10

This nep-for issue is ©2015 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.