nep-for New Economics Papers
on Forecasting
Issue of 2018‒04‒30
seven papers chosen by
Rob J Hyndman
Monash University

  1. Predicting the stock prices of G7 countries with Bitcoin prices By Afees A. Salisu; Kazeem Isah; Lateef O. Akanni
  2. A new approach for detecting shifts in forecast accuracy By Chiu, Ching-Wai (Jeremy); hayes, simon; kapetanios, george; Theodoridis, Konstantinos
  3. Forecasting Exchange Rates with Commodity Prices - A Global Country Analysis By Martin Baumgaertner; Jens Klose
  4. R2 bounds for predictive models: what univariate properties tell us about multivariate predictability By Stephen Wright; James Mitchell; Donald Robertson
  5. A Comparative Study of GARCH and EVT Model in Modeling Value-at-Risk By Li, Longqing
  6. “A regional perspective on the accuracy of machine learning forecasts of tourism demand based on data characteristics” By Oscar Claveria; Enric Monte; Salvador Torra
  7. Firm Performance and Macro Forecast Accuracy By Mari Tanaka; Nicholas Bloom; Maiko Koga; Haruko Kato

  1. By: Afees A. Salisu; Kazeem Isah (Centre for Econometric and Allied Research, University of Ibadan); Lateef O. Akanni (Department of Economics, University of Lagos,Akoka, Lagos, Nigeria)
    Abstract: This paper attempts to establish that some inherent features of the Bitcoin price can be exploited to produce better forecast results for stock prices. It does so by constructing predictive models for stock prices of G7 countries with symmetric and asymmetric prices of Bitcoin. The underlying statistical properties of Bitcoin prices such as persistence and conditional heteroscedasticity are captured in the estimation process using the Westerlund and Narayan (2015) estimator that allows for such effects in forecasting. There are two striking findings from the analysis. First, the results suggest that accounting for asymmetries is more likely to enhance the predictive power of Bitcoin in forecasting stock prices regardless of the data sample and forecast horizon. Secondly, the Bitcoin-based predictive model for stock prices, particularly the asymmetric variant, outperforms the Fractionally Integrated Autoregressive Moving Average (ARFIMA) model. While there are concerns as to whether the cryptocurrencies are veritable substitutes to the conventional financial assets, their close link with the developed stock exchanges such as those in the G7 countries suggests that they share some common characteristics such as news effects [asymmetries] which can be exploited when forecasting the behaviour of stock prices.
    Keywords: Stock price, Bitcoin price, G7 countries, Forecast evaluation
    JEL: C52 C53 G11 G14 G17
    Date: 2018–04
  2. By: Chiu, Ching-Wai (Jeremy) (Bank of England); hayes, simon (Bank of England); kapetanios, george (Kings College); Theodoridis, Konstantinos (Cardiff University)
    Abstract: Forecasts play a critical role at inflation-targeting central banks, such as the Bank of England. Breaks in the forecast performance of a model can potentially incur important policy costs. Commonly used statistical procedures, however, implicitly put a lot of weight on type I errors (or false positives), which result in a relatively low power of tests to identify forecast breakdowns in small samples. We develop a procedure which aims at capturing the policy cost of missing a break. We use data-based rules to find the test size that optimally trades off the costs associated with false positives with those that can result from a break going undetected for too long. In so doing, we also explicitly study forecast errors as a multivariate system. The covariance between forecast errors for different series, though often overlooked in the forecasting literature, not only enables us to consider testing in a multivariate setting but also increases the test power. As a result, we can tailor the choice of the critical values for each series not only to the in-sample properties of each series but also to how the series for forecast errors covary.
    Keywords: Forecast breaks; statistical decision making; central banking
    JEL: C53 E47 E58
    Date: 2018–04–13
  3. By: Martin Baumgaertner (THM Business School); Jens Klose (THM Business School)
    Abstract: This paper investigates the predictive properties of import and export prices of commodities on the exchange rates. A period from 1993 to 2016 is considered. We find that forecasts of the exchange rate adding commodity export and import prices are superior to those neglecting these variables. This holds irrespective of whether the countries are net exporters or importers of commodities. However, the forecasting power was even better in the 1990s and seems to have decreased since that that time. Nevertheless forecasts can even today be improved considerably by adding commodity prices.
    Keywords: Exchange Rate, Commodity Prices, Forecast, Panel-Analysis
    JEL: F17 F31 F47 C23
    Date: 2018
  4. By: Stephen Wright (Birkbeck, University of London); James Mitchell (Warwick Business School); Donald Robertson (University of Cambridge)
    Abstract: A longstanding puzzle in macroeconomic forecasting has been that a wide variety of multivariate models have struggled to out-predict univariate models consistently. We seek an explanation for this puzzle in terms of population properties. We derive bounds for the predictive R2 of the true, but unknown, multivariate model from univariate ARMA parameters alone. These bounds can be quite tight, implying little forecasting gain even if we knew the true multivariate model. We illustrate using CPI inflation data and the Eurozone in a specification motivated by a preferred-habitat model to test for monetary policy transmission domestically and internationally. Our findings suggest an impact of monetary policy on variance processes only and provides evidence for an international channel of monetary transmission on both money and capital markets. This is, to our knowledge, the first attempt to use search-engine data in the context of monetary policy.
    Keywords: attention, internet search, Google, monetary policy, ECB, FED, international financial markets, macro-finance, sovereign bonds, international finance, bond markets, preferred habitat models.
    JEL: C22 C32 C53 E37
    Date: 2018–04
  5. By: Li, Longqing
    Abstract: The paper addresses an inefficiency of the traditional approach in modeling the tail risk, particularly the 1-day ahead forecast of Value-at-Risk (VaR), using Extreme Value Theory (EVT) and GARCH model. Specifically, I apply both models onto major countries stock markets daily loss, including U.S., U.K., China and Hong Kong between 2006 and 2015, and compare the relative forecasting performance. The paper differs from other studies in two important ways. First, it incorporates an asymmetric shock of volatility in the financial time series. Second, it applies a skewed fat-tailed return distribution using the Generalized Error Distribution (GED). The back-testing result shows that, on one hand, the conditional EVT performs equally well relative to GARCH model under the Generalized Error Distribution. On the other hand, the Exponential GARCH based model is the best performing one in Value-at-Risk forecasting, because it not only correctly identifies the future extreme loss, but more importantly, its occurrence is independent.
    Keywords: Value-at-Risk,Extreme Value Theory, Backtesting, Risk Forecasting
    JEL: C53 C58 G32
    Date: 2017–02–25
  6. By: Oscar Claveria (AQR-IREA, University of Barcelona); Enric Monte (Polytechnic University of Catalunya (UPC)); Salvador Torra (Riskcenter-IREA, University of Barcelona)
    Abstract: In this work we assess the role of data characteristics in the accuracy of machine learning (ML) tourism forecasts from a spatial perspective. First, we apply a seasonal-trend decomposition procedure based on non-parametric regression to isolate the different components of the time series of international tourism demand to all Spanish regions. This approach allows us to compute a set of measures to describe the features of the data. Second, we analyse the performance of several ML models in a recursive multiple-step-ahead forecasting experiment. In a third step, we rank all seventeen regions according to their characteristics and the obtained forecasting performance, and use the rankings as the input for a multivariate analysis to evaluate the interactions between time series features and the accuracy of the predictions. By means of dimensionality reduction techniques we summarise all the information into two components and project all Spanish regions into perceptual maps. We find that entropy and dispersion show a negative relation with accuracy, while the effect of other data characteristics on forecast accuracy is heavily dependent on the forecast horizon.
    Keywords: STL decomposition; non-parametric regression; time series features; forecast accuracy; machine learning; tourism demand; regional analysis JEL classification: C45; C51; C53; C63; E27; L83
    Date: 2018–04
  7. By: Mari Tanaka (Hitotsubashi University, Graduate School of Economics); Nicholas Bloom (Stanford University, Department of Economics); Maiko Koga (Bank of Japan); Haruko Kato (Bank of Japan)
    Abstract: Ever since Keyenes' famous quote about animal spirits, there has been an interest in linking firms' expectations and actions. But the empirical evidence on this is scarce because of the lack of firm panel data on expectations and outcomes. In this paper, we combine a unique survey of Japanese firms' GDP forecasts with their accounting data for 27 years for over 1,000 large Japanese firms. We find four main results. First, we find that firms' GDP forecasts are positively and significantly associated with firms' input choices, such as investment and employment, and with firm's sales, even after controlling for year and firm fixed effects. These results are stronger for cyclical firms, suggesting a firm's input decision is particularly dependent on its manager's forecasts when its demand is more sensitive to the macro economy. Second, both optimistic and pessimistic forecast errors lower profitability because it is costly to have too much or too little capacity. Third, while over optimistic forecasts lower measured productivity, over pessimistic forecasts do not tend to have an impact on productivity. Finally, larger and more cyclical firms make more accurate forecasts, presumably reflecting the higher return from accurate forecasts. More productive, older, and bank owned firms also make more accurate forecasts, suggesting that forecasting ability is also linked to management ability, experience and governance. Collectively, this highlights the importance of firms' forecasting ability for micro and macro performance.
    Keywords: Forecast; investment; employment; productivity
    JEL: D22 D84
    Date: 2018–04–17

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.