ERN: Forecasting Techniques (Topic)最新文献

Predicting Returns Out of Sample: A Naïve Model Averaging Approach 预测样本外收益:Naïve模型平均方法

ERN: Forecasting Techniques (Topic) Pub Date : 2021-10-18 DOI: 10.2139/ssrn.3455866

Huafeng (Jason) Chen, Liang Jiang, Weiwei Liu

引用次数: 0

Forecasting with Deep Temporal Hierarchies 具有深层时间层次的预测

ERN: Forecasting Techniques (Topic) Pub Date : 2021-09-06 DOI: 10.2139/ssrn.3918315

Filotas Theodosiou, N. Kourentzes

{"title":"Forecasting with Deep Temporal Hierarchies","authors":"Filotas Theodosiou, N. Kourentzes","doi":"10.2139/ssrn.3918315","DOIUrl":"https://doi.org/10.2139/ssrn.3918315","url":null,"abstract":"In time series analysis and forecasting, the identification of an appropriate model remains a challenging task. Model misspecification can lead to erroneous forecasts and insights. The use of multiple views of the same time series by constructing temporally aggregate levels has been proposed as a way to overcome the model specification and selection uncertainty, with ample empirical evidence of forecast accuracy gains. Temporal Hierarchies is the most popular approach to achieve this, which itself is based on research in hierarchical forecasting. Although there has been substantial progress in this literature, the vast majority of methods rely on a restricted linear combination of different model outputs across the hierarchy. We investigate the use of deep learning to augment temporal hierarchies, relaxing the classical restrictions. Specifically, we look at deep learning for the generation of all the base forecasts, the hierarchical reconciliation, and an end-to-end method that embeds all steps in a single neural network. We inspect the performance of the proposed methods when applied to individual time series, or with global training across complete sets of series. We further investigate the requirements in terms of series set size, illustrating the conditions where deep learning temporal hierarchies outperform conventional temporal hierarchies.","PeriodicalId":170198,"journal":{"name":"ERN: Forecasting Techniques (Topic)","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116245713","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 4

Liquidity and Mispricing 流动性和错误定价

ERN: Forecasting Techniques (Topic) Pub Date : 2021-06-30 DOI: 10.2139/ssrn.3718411

D. Huber

引用次数: 1

Charting By Machines 机器制图

ERN: Forecasting Techniques (Topic) Pub Date : 2021-06-01 DOI: 10.2139/ssrn.3853436

Scott Murray, Houping Xiao, Yusen Xia

引用次数: 2

Lawrence R. Klein’s Principles in Modeling and Contributions in Nowcasting, Real-Time Forecasting, and Machine Learning Lawrence R. Klein的《建模原理及对临近预报、实时预测和机器学习的贡献》

ERN: Forecasting Techniques (Topic) Pub Date : 2020-09-29 DOI: 10.2139/ssrn.3702412

R. Mariano, Suleyman Ozmucur

{"title":"Lawrence R. Klein’s Principles in Modeling and Contributions in Nowcasting, Real-Time Forecasting, and Machine Learning","authors":"R. Mariano, Suleyman Ozmucur","doi":"10.2139/ssrn.3702412","DOIUrl":"https://doi.org/10.2139/ssrn.3702412","url":null,"abstract":"Lawrence R. Klein (September 14, 1920 – October 20, 2013), Nobel Laureate in Economic Sciences in 1980, was one of the leading figures in macro-econometric modeling. Although his contributions to forecasting using simultaneous equations macro models were very well known, his contributions to nowcasting and real-time forecasting, that he worked on in the last 30 years of his life, were generally overlooked by many researchers. The reasons for the miss are related to the ambiguity in terminology, specifically, the terms nowcast or nowcasting, and the empirical, though very significant, nature of his contributions. This paper reviews L. R. Klein’s guiding principles on modeling and his contributions to nowcasting and real-time forecasting, and discusses the connection of these contributions to the present state of fast evolving disciplines, such as economics, econometrics, statistics, data science, and machine learning. In so doing, we argue that L. R. Klein indeed expertly developed pioneering ideas and methodology for nowcasting and real-time forecasting; and the principles and contributions put forward by him are even more relevant now than ever.","PeriodicalId":170198,"journal":{"name":"ERN: Forecasting Techniques (Topic)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114561712","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

The Role of a Latent Value-Relevant Measure in Tracking and Predicting Stock Returns: A FAVAR Approach 潜在价值相关测度在股票收益追踪与预测中的作用:FAVAR方法

ERN: Forecasting Techniques (Topic) Pub Date : 2020-06-24 DOI: 10.2139/ssrn.3635073

Faisal M. Awwal, Xiaoquan Jiang

{"title":"The Role of a Latent Value-Relevant Measure in Tracking and Predicting Stock Returns: A FAVAR Approach","authors":"Faisal M. Awwal, Xiaoquan Jiang","doi":"10.2139/ssrn.3635073","DOIUrl":"https://doi.org/10.2139/ssrn.3635073","url":null,"abstract":"This paper attempts to estimate and study the role of 'other information', as posited in the residual income valuation model of Ohlson (1995), for tracking and predicting future returns of the S&P 500. 'Other information' is an unobserved variable and defined as a summary of value-relevant information about events and their effect on future profitability, which is captured in a company's current stock price and returns, but not yet reflected in a company's current financial statements. This suggests a potential to predict subsequent returns. Previous literature has found that traditional valuation metrics (e.g. B/P, E/P, and D/P ratios) have poor predictive power. In this study, we apply a factor augmented vector autoregression (FAVAR) to estimate this value-relevant latent variable and assess its predictive performance. The FAVAR is a suitable model because it enables us to analyze and quantify the linkages of stock market value, profitability, and unobserved factors that are broadly captured by big data. We use a two-step principal components estimation approach to extract the unobserved factors of 78 informational variables from financial market, accounting, investor and consumer sentiment, and macroeconomic data. Our analysis shows that, in comparison to competing measures, the estimated latent value-relevant variable can track contemporaneous stock returns and has statistically reliable power to predict both future real stock returns and excess returns over a Treasury Bill rate, both in- and out-of-sample.","PeriodicalId":170198,"journal":{"name":"ERN: Forecasting Techniques (Topic)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133263989","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

The Implied Cost of Capital: A Deep Learning Approach 隐含资本成本:一种深度学习方法

ERN: Forecasting Techniques (Topic) Pub Date : 2020-05-27 DOI: 10.2139/ssrn.3612472

Xinyu Wang

引用次数: 3

A Structural Model of Analyst Forecasts: Applications to Forecast Informativeness and Dispersion 分析师预测的结构模型:在预测信息量和分散性上的应用

ERN: Forecasting Techniques (Topic) Pub Date : 2020-05-18 DOI: 10.2139/ssrn.3604705

Jonathan Clarke, Soohun Kim, Kyuseok Lee, Kyoungwon Seo

引用次数: 0

Can Machine Learning Based Portfolios Outperform Traditional Risk-Based Portfolios? The Need to Account for Covariance Misspecification 基于机器学习的投资组合能胜过传统的基于风险的投资组合吗?需要解释协方差的错误说明

ERN: Forecasting Techniques (Topic) Pub Date : 2019-05-26 DOI: 10.3390/RISKS7030074

Prayut Jain, Shashi Jain

{"title":"Can Machine Learning Based Portfolios Outperform Traditional Risk-Based Portfolios? The Need to Account for Covariance Misspecification","authors":"Prayut Jain, Shashi Jain","doi":"10.3390/RISKS7030074","DOIUrl":"https://doi.org/10.3390/RISKS7030074","url":null,"abstract":"The Hierarchical risk parity (HRP) approach of portfolio allocation, introduced by Lopez de Prado (2016), applies graph theory and machine learning to build a diversified portfolio. Like the traditional risk-based allocation methods, HRP is also a function of the estimate of the covariance matrix, however, it does not require its invertibility. In this paper, we first study the impact of covariance misspecification on the performance of the different allocation methods. Next, we study under an appropriate covariance forecast model whether the machine learning based HRP outperforms the traditional risk-based portfolios. For our analysis, we use the test for superior predictive ability on out-of-sample portfolio performance, to determine whether the observed excess performance is significant or if it occurred by chance. We find that when the covariance estimates are crude, inverse volatility weighted portfolios are more robust, followed by the machine learning-based portfolios. Minimum variance and maximum diversification are most sensitive to covariance misspecification. HRP follows the middle ground; it is less sensitive to covariance misspecification when compared with minimum variance or maximum diversification portfolio, while it is not as robust as the inverse volatility weighed portfolio. We also study the impact of the different rebalancing horizon and how the portfolios compare against a market-capitalization weighted portfolio.","PeriodicalId":170198,"journal":{"name":"ERN: Forecasting Techniques (Topic)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130644519","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 21

Ensemble Models in Forecasting Financial Markets 金融市场预测中的集成模型

ERN: Forecasting Techniques (Topic) Pub Date : 2019-03-18 DOI: 10.21314/JCF.2019.374

Andreas S. Karathanasopoulos, M. Sovan, Chia Chun Lo, Adam Zaremba, Mohammed Osman

引用次数: 2