|
on Forecasting |
| By: | Shafiullah Qureshi (Department of Economics, Carleton University); Ba M Chu (Department of Economics, Carleton University); Fanny S. Demers (Department of Economics, Carleton University) |
| Abstract: | The objective of this paper is to apply state-of-the-art machine-learning (ML) algorithms to predict the monthly and quarterly real GDP growth of Canada using both Google trends (GT) and official data that are available ahead of the release of GDP data by Statistics Canada. This paper applies a novel approach for selecting features with XGBoost using the AutoML function of H2O. For this purpose, 5000 to 15000 XGBoost models are trained using this function. We use a very rigorous variable selection procedure, where only the best features are selected into the next stage to build a final learning model. Then pertinent features are introduced into the XGBoost model for forecasting real GDP growth rate. The forecasts are further improved by using Principal Component Analysis to choose the best factors out of the predictors selected by the XGBoost model. The results indicate that there are gains in nowcasting accuracy from using the XGBoost model with this two- step strategy. We first find that XGBoost is a superior forecasting model relative to our baseline models using alternative forecasting procedures such as AR(1). We also find that Google Trends data combined with the XGBoost model provide a very viable source of information for predicting Canadian real GDP growth when official data are not yet available due to publication lags. Thus, we can forecast real GDP growth rate accurately ahead of the release of official data. Moreover, we apply various techniques to make the machine learning model more interpretable. |
| Date: | 2020–08 |
| URL: | http://d.repec.org/n?u=RePEc:car:carecp:20-14&r=all |
| By: | Glocker, Christian; Kaniovski, Serguei |
| Abstract: | We propose a modeling approach involving a series of small-scale dynamic factor models. They are connected to each other within a cluster, whose linkages are derived from Granger-causality tests. This approach merges the benefits of large-scale macroeconomic and small-scale factor models, rendering our Cluster of Dynamic Factor Models (CDFM) useful for model-consistent nowcasting and forecasting on a larger scale. While the CDFM has a simple structure and is easy to replicate, its forecasts are more precise than those of a wide range of competing models and those of professional forecasters. Moreover, the CDFM allows forecasters to introduce their own judgment and hence produce conditional forecasts. |
| Keywords: | Forecasting, Dynamic factor model, Granger causality, Structural modeling |
| JEL: | C22 C53 C55 E37 |
| Date: | 2020–07 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:101874&r=all |
| By: | Ryan Cumings-Menon; Minchul Shin |
| Abstract: | We propose probability and density forecast combination methods that are defined using the entropy regularized Wasserstein distance. First, we provide a theoretical characterization of the combined density forecast based on the regularized Wasserstein distance under the Gaus-sian assumption. Second, we show how this type of regularization can improve the predictive power of the resulting combined density. Third, we provide a method for choosing the tuning parameter that governs the strength of regularization. Lastly, we apply our proposed method to the U.S. inflation rate density forecasting, and illustrate how the entropy regularization can improve the quality of predictive density relative to its unregularized counterpart. |
| Keywords: | Entropy regularization; Wasserstein distance; optimal transport; density fore-casting; model combination. |
| JEL: | C53 E37 |
| Date: | 2020–08–06 |
| URL: | http://d.repec.org/n?u=RePEc:fip:fedpwp:88545&r=all |
| By: | Fantazzini, Dean |
| Abstract: | The ability of Google Trends data to forecast the number of new daily cases and deaths of COVID-19 is examined using a dataset of 158 countries. The analysis includes the computations of lag correlations between confirmed cases and Google data, Granger causality tests, and an out-of-sample forecasting exercise with 18 competing models with a forecast horizon of 14 days ahead. This evidence shows that Google-augmented models outperform the competing models for most of the countries. This is significant because Google data can complement epidemiological models during difficult times like the ongoing COVID-19 pandemic, when official statistics maybe not fully reliable and/or published with a delay. Moreover, real-time tracking with online-data is one of the instruments that can be used to keep the situation under control when national lockdowns are lifted and economies gradually reopen. |
| Keywords: | Covid-19; Google Trends; VAR; ARIMA; ARIMA-X; ETS; LASSO; SIR model |
| JEL: | C22 C32 C51 C53 G17 I18 I19 |
| Date: | 2020–08 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:102315&r=all |
| By: | L. Vanessa Smith (DERS, University of York.); Nori Tarui (Department of Economics, University of Hawaii.); Takashi Yamagata (DERS, University of York & ISER, Osaka University.) |
| Abstract: | This paper demonstrates how the global vector autoregressive (GVAR) modelling framework can be used for producing conditional forecasts of global fossil fuel consumption and CO2 emissions, as well as for conducting counterfactual analysis related to carbon pricing, conditional on alternative GDP scenarios. The choice of the conditioning variable does not limit the generality of the approach. The proposed analysis can be useful in guiding and informing policy making as illustrated by our application, which conditions on two-year horizon GDP forecast trajectories by the International Monetary Fund. These trajectories are associated with the global economic shock due to the COVID-19 pandemic. Our model makes use of a unique quarterly data set of coal, natural gas, and oil consumption, output and exchange rates, including global fossil fuel prices for 32 major CO2 emitting countries. The results show that fossil fuel consumption and CO2 emissions are expected to return to their pre-crisis levels, and even exceed them, within the two-year horizon despite the large reductions in the first quarter following the outbreak. More robust growth is anticipated for emerging than for advanced economies. Recovery to the pre-crisis levels is expected even if another wave of pandemic occurs within a year. Results from the counterfactual carbon pricing scenario indicate that an increase in coal prices is expected to have a smaller impact on GDP than on fossil fuel consumption. Thus, the COVID-19 pandemic would not provide countries with a strong reason to delay climate change mitigation efforts. |
| Keywords: | fuel consumption; CO2 emissions; Global VAR (GVAR); conditional forecasts; carbon pricing; COVID-19 |
| JEL: | C33 O50 P18 Q41 Q43 Q47 |
| Date: | 2020–08 |
| URL: | http://d.repec.org/n?u=RePEc:was:dpaper:2004&r=all |
| By: | Peng-Fei Dai (TJU); Xiong Xiong (TJU); Wei-Xing Zhou (ECUST) |
| Abstract: | This paper aims to examine whether the global economic policy uncertainty (GEPU) and uncertainty changes have different impacts on crude oil futures volatility. We establish single-factor and two-factor models under the GARCH-MIDAS framework to investigate the predictive power of GEPU and GEPU changes excluding and including realized volatility. The findings show that the models with rolling-window specification perform better than those with fixed-span specification. For single-factor models, the GEPU index and its changes, as well as realized volatility, are consistent effective factors in predicting the volatility of crude oil futures. Specially, GEPU changes have stronger predictive power than the GEPU index. For two-factor models, GEPU is not an effective forecast factor for the volatility of WTI crude oil futures or Brent crude oil futures. The two-factor model with GEPU changes contains more information and exhibits stronger forecasting ability for crude oil futures market volatility than the single-factor models. The GEPU changes are indeed the main source of long-term volatility of the crude oil futures. |
| Date: | 2020–07 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2007.12838&r=all |
| By: | Claudia Foroni; Francesco Ravazzolo; Luca Rossini |
| Abstract: | We analyse the importance of low frequency hard and soft macroeconomic information, respectively the industrial production index and the manufacturing Purchasing Managers' Index surveys, for forecasting high-frequency daily electricity prices in two of the main European markets, Germany and Italy. We do that by means of mixed-frequency models, introducing a Bayesian approach to reverse unrestricted MIDAS models (RU-MIDAS). Despite the general parsimonious structure of standard MIDAS models, the RU-MIDAS has a large set of parameters when several predictors are considered simultaneously and Bayesian inference is useful for imposing parameter restrictions. We study the forecasting accuracy for different horizons (from $1$ day ahead to $28$ days ahead) and by considering different specifications of the models. Results indicate that the macroeconomic low frequency variables are more important for short horizons than for longer horizons. Moreover, accuracy increases by combining hard and soft information, and using only surveys gives less accurate forecasts than using only industrial production data. |
| Date: | 2020–07 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2007.13566&r=all |
| By: | Cimadomo, Jacopo; Giannone, Domenico; Lenza, Michele; Sokol, Andrej; Monti, Francesca |
| Abstract: | Monitoring economic conditions in real time, or nowcasting, is among the key tasks routinely performed by economists. Nowcasting entails some key challenges, which also characterise modern Big Data analytics, often referred to as the three \Vs": the large number of time series continuously released (Volume), the complexity of the data covering various sectors of the economy, published in an asynchronous way and with different frequencies and precision (Variety), and the need to incorporate new information within minutes of their release (Velocity). In this paper, we explore alternative routes to bring Bayesian Vector Autoregressive (BVAR) models up to these challenges. We find that BVARs are able to effectively handle the three Vs and produce, in real time, accurate probabilistic predictions of US economic activity and, in addition, a meaningful narrative by means of scenario analysis. JEL Classification: E32, E37, C01, C33, C53 |
| Keywords: | Big Data, business cycles, forecasting, mixed frequencies, real time, scenario analysis |
| Date: | 2020–08 |
| URL: | http://d.repec.org/n?u=RePEc:ecb:ecbwps:20202453&r=all |