M5 competition uncertainty: Overdispersion,distributional forecasting,GAMLSS, and beyond

The M5 competition uncertainty track aims for probabilistic forecasting of sales of thousands of Walmart retail goods. We show that the M5 competition data face strong overdispersion and sporadic demand, especially zero demand. We discuss modeling issues concerning adequate probabilistic forecasting of such count data processes. Unfortunately, the majority of popular prediction methods used in the M5 competition (e.g. lightgbm and xgboost GBMs) fail to address the data characteristics, due to the considered objective functions. Distributional forecasting provides a suitable modeling approach to overcome those problems. The GAMLSS framework allows for flexible probabilistic forecasting using low-dimensional distributions. We illustrate how the GAMLSS approach can be applied to M5 competition data by modeling the location and scale parameters of various distributions, e.g. the negative binomial distribution. Finally, we discuss software packages for distributional modeling and their drawbacks, like the R package gamlss with its package extensions, and (deep) distributional forecasting libraries such as TensorFlow Probability.     

Understanding machine learning-based forecasting methods: A decomposition framework and research opportunities

Machine learning (ML) methods are gaining popularity in the forecasting field, as they have shown strong empirical performance in the recent M4 and M5 competitions, as well as in several Kaggle competitions. However, understanding why and how these methods work well for forecasting is still at a very early stage, partly due to their complexity. In this paper, I present a framework for regression-based ML that provides researchers with a common language and abstraction to aid in their study. To demonstrate the utility of the framework, I show how it can be used to map and compare ML methods used in the M5 Uncertainty competition. I then describe how the framework can be used together with ablation testing to systematically study their performance. Lastly, I use the framework to provide an overview of the solution space in regression-based ML forecasting, identifying areas for further research.     

Evaluating quantile forecasts in the M5 uncertainty competition

Probabilistic forecasts are necessary for robust decisions in the face of uncertainty. The M5 Uncertainty competition required participating teams to forecast nine quantiles for unit sales of various products at various aggregation levels and for different time horizons. This paper evaluates the forecasting performance of the quantile forecasts at different aggregation levels and at different quantile levels. We contrast this with some theoretical predictions, and discuss potential implications and promising future research directions for the practice of probabilistic forecasting.     

Board internationalization and corporate social responsibility

This study examines the association between board internationalization and firms’ corporate social responsibility (CSR) performance in China during 2009–2019. Using a large dataset of director nationalities and CSR scores, we find that foreign directors promote CSR performance as evidenced by higher CSR scores. We further document that this relationship is more pronounced among government-controlled firms, firms with better corporate governance, and firms operating under stricter institutional environments. These findings remain valid after sequential robustness checks. Overall, our findings imply that foreign directors motivate the board of directors and their firms to actively pursue and practice social responsibility. Our study enriches the literature on the outcomes of board internationalization and determinants of CSR and provides the internationalization of corporate governance mechanisms a reasonable basis.     

The M4 Competition: Results,findings, conclusion and way forward

The M4 competition is the continuation of three previous competitions started more than 45 years ago whose purpose was to learn how to improve forecasting accuracy, and how such learning can be applied to advance the theory and practice of forecasting. The purpose of M4 was to replicate the results of the previous ones and extend them into three directions: First significantly increase the number of series, second include Machine Learning (ML) forecasting methods, and third evaluate both point forecasts and prediction intervals. The five major findings of the M4 Competitions are: 1. Out Of the 17 most accurate methods, 12 were “combinations” of mostly statistical approaches. 2. The biggest surprise was a “hybrid” approach that utilized both statistical and ML features. This method’s average sMAPE was close to 10% more accurate than the combination benchmark used to compare the submitted methods. 3. The second most accurate method was a combination of seven statistical methods and one ML one, with the weights for the averaging being calculated by a ML algorithm that was trained to minimize the forecasting. 4. The two most accurate methods also achieved an amazing success in specifying the 95% prediction intervals correctly. 5. The six pure ML methods performed poorly, with none of them being more accurate than the combination benchmark and only one being more accurate than Naïve2. This paper presents some initial results of M4, its major findings and a logical conclusion. Finally, it outlines what the authors consider to be the way forward for the field of forecasting.     

Does ethics improve stock market resilience in times of instability?

This paper compares the resilience of ethical (Islamic and socially responsible) indexes among five developed (US, UK, Japan, Canada, Australia) and three emerging markets (Brazil, India, South Africa) during the period following the 2008 subprime crisis. It relies on a multivariate CAPM-EGARCH model that accounts for sudden changes in volatility through the application of an iterated cumulative sums of squares (ICSS) algorithm on daily data over the sample period 2008–2014 to model time-varying volatility and ensure reliable estimates. The study confirms the lower systemic risk associated with Islamic indexes during the bearish period and reports that SRI, despite being more subject to systemic risk, offered higher alphas in highly integrated markets, while Islamic indexes performed better in less integrated ones. The evidence also reveals a very limited increase in the models’ predictability power from the integration of sudden changes in volatility into the EGARCH models during the full sample period. This limit is more marked during the bearish sub-period. Our findings have important implications for international investment and portfolio diversification perspectives in times of financial downturn.     

Telling their own stories: Māori entrepreneurship in the mainstream screen industry

We examined how factors from Indigenous entrepreneurship research (social capital, cultural capital, self-efficacy) help explain the high level of Māori entrepreneurial performance in the mainstream screen industry. Results, based on ten case studies and a one-year series of structured interviews, extend prior research by showing that these Indigenous entrepreneurs benefit jointly from two forms of capital: cultural and social. We found high levels of both forms to increase the desire for emancipation of cultural and community identity – not just individual identity – through entrepreneurship. Self-efficacy and storytelling helped ameliorate discontinuities across Indigenous and mainstream contexts. Our research sheds new light on how Indigenous ventures can pursue mainstream entrepreneurship while maintaining cultural identity. It also makes several distinct contributions to the Indigenous entrepreneurship literature. First, it provides an integrative theoretic review. Second, it illustrates a culturally appropriate methodology for researching Māori entrepreneurs with implications for other Indigenous communities. Third, it proposes cultural capital and social capital as a two-part framework for explaining Indigenous entrepreneurial action. Fourth, it shows how entrepreneurship can be empowering for Indigenous communities. Finally, our paper demonstrates that entrepreneurship is a promising mechanism for preserving and promoting the cultures of Māori and other Indigenous peoples.     

The trade-off between innovation and defense industrial policy: A simulation model analysis of the Norwegian defense industry

The paper investigates the trade-off between innovation and defense industrial policy. It presents an agent-based simulation model calibrated for the Norwegian defense industry that compares different policy scenarios and examines the effects of a pending EU market liberalization process. The paper points to two main results. (1) It finds that a pure scenario where national authorities focus on, and provide support exclusively for, either a) international competitiveness or b) national defense and security objectives, is more Pareto efficient than a corresponding mixed strategy where policy makers simultaneously pursue both international competitiveness and defense and security objectives. (2) Under the conditions of the new EU liberalization regime, it finds that a stronger and more visible trade-off will emerge between international competitiveness and national defense and security objectives. Policy makers will have to choose which to prioritize, and set a clear agenda focusing on one of the two objectives.     

基于BP神经网络的我国M2/GDP预测与对策

本文计算了1993年至2005年我国M2/GDP,发现该数据总体呈上升趋势,这一定程度上反映了我国体制上的深层次问题。另外,本文运用带有动量项和自适应学习率的BP神经网络方法预测出我国2006年的M2/GDP是1.6386左右,并提出了一些相应的政策建议。     

贷款行业集中度对商业银行系统性风险影响的实证研究——基于MES方法

金融危机爆发后,银行业系统性风险得到了广泛的关注。基于2007-2012年中国10家上市银行的相关数据,文章利用MES方法测算系统性风险贡献度,分析贷款行业集中度对商业银行系统性风险的影响。研究结果表明,贷款行业集中度对商业银行系统性风险存在正向影响。因此,商业银行在投放贷款过程中,应该防范贷款行业投向的过度集中,从而减少系统性风险爆发的可能性。该研究结论为我国商业银行合理配置信贷资源提供了有利的依据。