Journal of Forecasting最新文献

{"title":"Regime-Switching Density Forecasts Using Economists' Scenarios","authors":"Graziano Moramarco","doi":"10.1002/for.3228","DOIUrl":"https://doi.org/10.1002/for.3228","url":null,"abstract":"<p>We propose an approach for generating macroeconomic density forecasts that incorporate information on multiple scenarios defined by experts. We adopt a regime-switching framework in which sets of scenarios (“views”) are used as Bayesian priors on economic regimes. Predictive densities coming from different views are then combined by optimizing objective functions of density forecasting. We illustrate the approach with an empirical application to quarterly real-time forecasts of the US GDP growth rate, in which we exploit the Fed's macroeconomic scenarios used for bank stress tests. We show that the approach achieves good accuracy in terms of average predictive scores and good calibration of forecast distributions. Moreover, it can be used to evaluate the contribution of economists' scenarios to density forecast performance.</p>","PeriodicalId":47835,"journal":{"name":"Journal of Forecasting","volume":"44 2","pages":"833-845"},"PeriodicalIF":3.4,"publicationDate":"2024-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/for.3228","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143119266","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Using a Wage–Price-Setting Model to Forecast US Inflation","authors":"Nguyen Duc Do","doi":"10.1002/for.3210","DOIUrl":"https://doi.org/10.1002/for.3210","url":null,"abstract":"<div>\u0000 \u0000 <p>This study modifies a wage–price-setting (WPS) model to forecast US inflation over 1- to 3-year horizons, based on the assumption that firms use a rule of thumb to set prices after settling a wage agreement. The out-of-sample forecast results show that productivity growth is a powerful predictor of inflation, in the sense that during the 1990Q1–2023Q4 period, the modified WPS model improved upon some univariate benchmark models and multivariate models such as the Phillips curve, term spread, and wage-inflation models. From the early 2000s to the prepandemic period, forecast accuracy was improved by combining productivity growth with anchored inflation expectations. Interestingly, during this period, forecasts derived from the WPS model with constant-inflation expectations were found to slightly outperform Greenbook forecasts in forecasting quarter-over-quarter inflation from two- to four-quarter horizons.</p>\u0000 </div>","PeriodicalId":47835,"journal":{"name":"Journal of Forecasting","volume":"44 2","pages":"803-832"},"PeriodicalIF":3.4,"publicationDate":"2024-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143118143","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Quantification Approach of Changes in Firms' Financial Situation Using Neural Networks for Predicting Bankruptcy","authors":"Philippe du Jardin","doi":"10.1002/for.3227","DOIUrl":"https://doi.org/10.1002/for.3227","url":null,"abstract":"<div>\u0000 \u0000 <p>For a very long time, bankruptcy models were considered ahistorical, as they were mostly based on ratios measured over a single year. However, time is an essential variable that explains a firm's ability to survive. It is precisely for these reasons that measures intended to represent firm history have been studied and progressively used to complement traditional explanatory variables using financial ratios or variation indicators of such ratios. Even if these measures are not totally useless, they failed to be widely used in the literature. This is the reason why we propose a method, called temporal financial pattern–based method (TPM) that makes it possible to efficiently represent a firm's history using a quantification process and use the result of this process to improve model accuracy. This method relies on an estimation of typical temporal financial patterns that govern changes in a firm's financial situation over time, using neural networks. The results demonstrate that TPM leads to better prediction accuracy than that achieved with traditional models.</p>\u0000 </div>","PeriodicalId":47835,"journal":{"name":"Journal of Forecasting","volume":"44 2","pages":"781-802"},"PeriodicalIF":3.4,"publicationDate":"2024-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143117895","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Global Risk Aversion: Driving Force of Future Real Economic Activity","authors":"Jinhwan Kim,&nbsp;Hoon Cho,&nbsp;Doojin Ryu","doi":"10.1002/for.3203","DOIUrl":"https://doi.org/10.1002/for.3203","url":null,"abstract":"<p>This study examines how global risk aversion affects future real economic activity (REA). We propose a new international real business cycle (RBC) framework with a stochastic global risk aversion spillover process by extending the RBC model. Our model suggests output competition and risk aversion spillover as two influence channels of global risk aversion. We extract relative risk aversion factors and evaluate the significance of changes in the level of global risk aversion for forecasting. Our findings suggest that changes in the level of global risk aversion significantly drive the business cycle of open economies. A global risk aversion factor predicts a domestic country's future REA at least as well as the domestic risk aversion factor does. The impact of global risk aversion can vary depending on a country's relative productivity.</p>","PeriodicalId":47835,"journal":{"name":"Journal of Forecasting","volume":"44 2","pages":"706-729"},"PeriodicalIF":3.4,"publicationDate":"2024-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/for.3203","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143117892","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Data-Driven Predictive Modeling of Citywide Crowd Flow for Urban Safety Management: A Case Study of Beijing, China","authors":"He Jiang,&nbsp;Xuxilu Zhang,&nbsp;Yao Dong,&nbsp;Jianzhou Wang","doi":"10.1002/for.3216","DOIUrl":"https://doi.org/10.1002/for.3216","url":null,"abstract":"&lt;div&gt;\u0000 \u0000 &lt;p&gt;Crowd flow forecasting is vital for urban planning, resource allocation, and public safety, particularly in the context of the COVID-19 pandemic. However, traditional predictive models struggle to capture the complex, nonlinear spatial–temporal relationships inherent in crowd flow data due to its irregular volatility. To address these limitations, this paper proposes the innovative citywide crowd flow prediction (CCFP) model, which merges statistical rules with machine learning techniques (XGBoost, LightGBM, and CatBoost). The CCFP model is specifically designed to leverage spatial dependencies and two-level periodicity (weekly and daily) in population flow to predict crowd flow indexes (\u0000&lt;span&gt;&lt;/span&gt;&lt;math&gt;\u0000 &lt;semantics&gt;\u0000 &lt;mrow&gt;\u0000 &lt;mi&gt;C&lt;/mi&gt;\u0000 &lt;mi&gt;F&lt;/mi&gt;\u0000 &lt;mi&gt;I&lt;/mi&gt;\u0000 &lt;/mrow&gt;\u0000 &lt;annotation&gt;$$ CFI $$&lt;/annotation&gt;\u0000 &lt;/semantics&gt;&lt;/math&gt;) within specific areas. We employ an urban area graph created using the Node2Vec algorithm to capture the temporal and spatial nuances of human flow patterns. Notably, this study innovatively incorporates migration, weather, and epidemic data into machine-learning models for feature extraction. Moreover, it introduces weighted factors—\u0000&lt;span&gt;&lt;/span&gt;&lt;math&gt;\u0000 &lt;semantics&gt;\u0000 &lt;mrow&gt;\u0000 &lt;mi&gt;g&lt;/mi&gt;\u0000 &lt;mi&gt;r&lt;/mi&gt;\u0000 &lt;mi&gt;o&lt;/mi&gt;\u0000 &lt;mi&gt;w&lt;/mi&gt;\u0000 &lt;mi&gt;t&lt;/mi&gt;\u0000 &lt;mi&gt;h&lt;/mi&gt;\u0000 &lt;mo&gt;,&lt;/mo&gt;\u0000 &lt;mspace&gt;&lt;/mspace&gt;\u0000 &lt;mi&gt;b&lt;/mi&gt;\u0000 &lt;mi&gt;a&lt;/mi&gt;\u0000 &lt;mi&gt;s&lt;/mi&gt;\u0000 &lt;mi&gt;e&lt;/mi&gt;\u0000 &lt;mo&gt;,&lt;/mo&gt;\u0000 &lt;mspace&gt;&lt;/mspace&gt;\u0000 &lt;mi&gt;w&lt;/mi&gt;\u0000 &lt;mi&gt;e&lt;/mi&gt;\u0000 &lt;mi&gt;e&lt;/mi&gt;\u0000 &lt;mi&gt;k&lt;/mi&gt;\u0000 &lt;/mrow&gt;\u0000 &lt;annotation&gt;$$ growth, base, week $$&lt;/annotation&gt;\u0000 &lt;/semantics&gt;&lt;/math&gt;, and \u0000&lt;span&gt;&lt;/span&gt;&lt;math&gt;\u0000 &lt;semantics&gt;\u0000 &lt;mrow&gt;\u0000 &lt;mi&gt;d&lt;/mi&gt;\u0000 &lt;mi&gt;a&lt;/mi&gt;\u0000 &lt;mi&gt;y&lt;/mi&gt;\u0000 &lt;/mrow&gt;\u0000 &lt;annotation&gt;$$ day $$&lt;/annotation&gt;\u0000 &lt;/semantics&gt;&lt;/math&gt;—to enhance the accuracy of \u0000&lt;span&gt;&lt;/span&gt;&lt;math&gt;\u0000 &lt;semantics&gt;\u0000 &lt;mrow&gt;\u0000 &lt;mi&gt;C&lt;/mi&gt;\u0000 &lt;mi&gt;F&lt;/mi&gt;\u0000 &lt;mi&gt;I&lt;/mi&gt;\u0000 &lt;/mrow&gt;\u0000 &lt;annotation&gt;$$ CFI $$&lt;/annotation&gt;\u0000 &lt;/semantics&gt;&lt;/math&gt; prediction. Among the combined models, CCFP outperforms others with remarkable scientific precision (root mean squared error, \u0000&lt;span&gt;&lt;/span&gt;&lt;math&gt;\u0000 &lt;semantics&gt;\u0000 &lt;mrow&gt;\u0000 &lt;mi&gt;R&lt;/mi&gt;\u0000 &lt;mi&gt;M&lt;/mi&gt;\u0000 &lt;mi&gt;S&lt;/mi&gt;\u0000 &lt;mi&gt;E&lt;/mi&gt;\u0000 &lt;mo&gt;=&lt;/mo&gt;\u0000 &lt;mn&gt;2.04&lt;/mn&gt;\u0000 &lt;/mrow&gt;\u0000 ","PeriodicalId":47835,"journal":{"name":"Journal of Forecasting","volume":"44 2","pages":"730-752"},"PeriodicalIF":3.4,"publicationDate":"2024-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143117893","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cross-Learning With Panel Data Modeling for Stacking and Forecast Time Series Employment in Europe","authors":"Pietro Giorgio Lovaglio","doi":"10.1002/for.3224","DOIUrl":"https://doi.org/10.1002/for.3224","url":null,"abstract":"<p>This paper describes the use of cross-learning with panel data modeling for stacking regressions of different predictive models for time series employment across occupations in Europe during the last 15 years. The ARIMA and state space models were used for the predictions on the first-level model ensemble. On the second level, the time series predictions of these models were combined for stacking, using panel data estimators as a cross-learner and also exploiting the strong hierarchical data structure (time series nested in occupational groups). Very few methods adopt stacking to generate ensembles for time series regressions. Indeed, to the best of our knowledge, panel data modeling has never before been used as a cross-learner in staking strategies. Empirical application was used to fit employment by occupations in 30 European countries between 2010 Q1 and 2022 Q4, using the last year as the test set. The empirical results show that using panel data modeling as a multivariate time series cross-learner that stacks univariate time series base models—especially when they do not produce accurate predictions—is an alternative worthy of consideration, also with respect to such classical aggregation schemes as optimal and equal weighting.</p>","PeriodicalId":47835,"journal":{"name":"Journal of Forecasting","volume":"44 2","pages":"753-780"},"PeriodicalIF":3.4,"publicationDate":"2024-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/for.3224","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143117894","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Forecasting Realized Volatility: The Choice of Window Size","authors":"Yuqing Feng,&nbsp;Yaojie Zhang","doi":"10.1002/for.3221","DOIUrl":"https://doi.org/10.1002/for.3221","url":null,"abstract":"<div>\u0000 \u0000 <p>Different window sizes may produce different empirical results. However, how to choose an ideal window size is still an open question. We investigate how the window size affects the predictive performance of volatility. The empirical results show that the loss function for volatility prediction takes on a U-shape as the window size increases. This suggests that if the window size is chosen too large or too small, the loss function tends to be large and the model's predictive accuracy decreases. A window size of between 1000 and 2000 observations is ideal for various assets because it can produce relatively minimal forecast errors. From an asset allocation perspective, a mean–variance investor can obtain sizeable utility by using a model with a low loss function value for her portfolio. Moreover, the results are robust in a variety of settings.</p>\u0000 </div>","PeriodicalId":47835,"journal":{"name":"Journal of Forecasting","volume":"44 2","pages":"692-705"},"PeriodicalIF":3.4,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143117165","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Taming Data-Driven Probability Distributions","authors":"Jozef Baruník,&nbsp;Luboš Hanus","doi":"10.1002/for.3208","DOIUrl":"https://doi.org/10.1002/for.3208","url":null,"abstract":"<p>We propose a deep learning approach to probabilistic forecasting of macroeconomic and financial time series. By allowing complex time series patterns to be learned from a data-rich environment, our approach is useful for decision making that depends on the uncertainty of a large number of economic outcomes. In particular, it is informative for agents facing asymmetric dependence of their loss on the outcomes of possibly non-Gaussian and nonlinear variables. We demonstrate the usefulness of the proposed approach on two different datasets where a machine learns patterns from the data. First, we illustrate the gains in predicting stock return distributions that are heavy tailed and asymmetric. Second, we construct macroeconomic fan charts that reflect information from a high-dimensional dataset.</p>","PeriodicalId":47835,"journal":{"name":"Journal of Forecasting","volume":"44 2","pages":"676-691"},"PeriodicalIF":3.4,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/for.3208","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143117166","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ensemble Multitask Prediction of Air Pollutants Time Series: Based on Variational Inference, Data Projection, and Generative Adversarial Network","authors":"Kang Wang,&nbsp;Chao Qu,&nbsp;Jianzhou Wang,&nbsp;Zhiwu Li,&nbsp;Haiyan Lu","doi":"10.1002/for.3218","DOIUrl":"https://doi.org/10.1002/for.3218","url":null,"abstract":"<div>\u0000 \u0000 <p>In light of the mounting environmental pressures, especially the significant threat urban air pollution poses to public health, there arises an imperative need to develop a data-driven model for air pollution prediction. However, contemporary deep learning techniques, such as recurrent neural networks, often struggle to effectively capture the underlying data patterns and distributions, resulting in reduced model stability. To address this gap, this study introduces an ensemble Wasserstein generative adversarial network framework (EWGF) to enhance the stability and accuracy of PM_2.5 predictions by facilitating the acquisition of more informative data representations through Wasserstein generative adversarial network. The framework contains an intricate feature extraction pipeline that automatically learns features containing residual information as representations of potential features, effectively ameliorating the underutilization of feature information. We address a nonconvex multi-objective optimization problem associated with amalgamating diverse Wasserstein generative adversarial network architectures, which enhance the inherent instability of the predictions. Furthermore, an adaptive search strategy is introduced to ascertain the optimal distribution of prediction residuals, thereby expanding the prediction interval estimation method based on residual distribution. We rigorously evaluate the proposed framework using datasets from three major Indian cities, and our experiments unequivocally show that the EWGF outperforms existing solutions in both PM_2.5 point prediction and interval prediction, evidenced by an approximate 8.07% reduction in mean absolute percentage error and an approximate 19.41% improvement in prediction interval score compared to the baseline model.</p>\u0000 </div>","PeriodicalId":47835,"journal":{"name":"Journal of Forecasting","volume":"44 2","pages":"646-675"},"PeriodicalIF":3.4,"publicationDate":"2024-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143116411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Vector SHAP Values for Machine Learning Time Series Forecasting","authors":"Ji Eun Choi,&nbsp;Ji Won Shin,&nbsp;Dong Wan Shin","doi":"10.1002/for.3220","DOIUrl":"https://doi.org/10.1002/for.3220","url":null,"abstract":"<div>\u0000 \u0000 <p>We propose a new vector SHapley Additive exPlanations (SHAP) to interpret machine learning models for forecasting time series using lags of predictor variables. Unlike the standard SHAP measuring the contribution of each lag of each predictor variable, the proposed vector SHAP measures the contribution of the vector of the lags of each variable. The vector SHAP has an advantage of faster computation over the standard SHAP. Some desirable properties of the vector SHAP (vector local accuracy, vector missingness, and vector consistency) are established. A Monte Carlo simulation shows that the vector SHAP has a much faster computing time than the SHAP; the difference of the standard SHAP and the vector SHAP is small; the sampling SHAP is sensitive to the sampling proportion in a range of practical application; the vector SHAP mitigates the sensitivity issue. The vector SHAP is applied to the realized volatility of world major stock price indices of 16 countries for forecasting the realized volatility of South Korea stock price index, KOSPI. Further vectoring by regions of Europe, North America, and Asia yields vector SHAP value for each region which is very close to the sum of vector SHAP values of the countries of the region, illustrating usefulness of the strategy of vectoring.</p>\u0000 </div>","PeriodicalId":47835,"journal":{"name":"Journal of Forecasting","volume":"44 2","pages":"635-645"},"PeriodicalIF":3.4,"publicationDate":"2024-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143116444","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}