nep-big New Economics Papers
on Big Data
Issue of 2019‒11‒11
eleven papers chosen by
Tom Coupé
University of Canterbury

  1. Industrial Growth in Sub-Saharan Africa: Evidence from Machine Learning with Insights from Nightlight Satellite Images By Christian S. Otchia; Simplice A. Asongu
  2. Artificial Intelligence in Education: Can AI bring the full potential of personalized learning to education? By van der Vorst, Tommy; Jelicic, Nick
  3. Deep Learning for Stock Selection Based on High Frequency Price-Volume Data By Junming Yang; Yaoqi Li; Xuanyu Chen; Jiahang Cao; Kangkang Jiang
  4. The survival of start-ups in time of crisis. A machine learning approach to measure innovation By Marco Guerzoni; Consuelo R. Nava; Massimiliano Nuccio
  5. Business opportunities in 5G mobile technology By Nyström, Anna-Greta; Gugenishvili, Ilia
  6. Estimating Freight Transport Activity Using Nighttime Lights Satellite Data in China, India and Saudi Arabia By Hector G. Lopez-Ruiz; Nora Nezamuddin; Reema Al Hassan; Abdelrahman Muhsen
  7. The transport-based meshfree method (TMM) and its applications in finance: a review By Philippe G. LeFloch; Jean-Marc Mercier
  8. The impact of AI on employment: a historical account of its evolution By Garcia-Murillo, Martha; MacInnes, Ian
  9. Deep Hedging: Learning to Simulate Equity Option Markets By Magnus Wiese; Lianjun Bai; Ben Wood; Hans Buehler
  10. Artificial Intelligence: Governments Large and Small Take on the Issues By Longstaff, P.H.
  11. Demand Analysis with Many Prices By Victor Chernozhukov; Jerry A. Hausman; Whitney K. Newey

  1. By: Christian S. Otchia (Kwansei Gakuin University, Japan); Simplice A. Asongu (Yaoundé, Cameroon)
    Abstract: This study uses nightlight time data and machine learning techniques to predict industrial development in Africa. The results provide the first evidence on how machine learning techniques and nightlight data can be used to predict economic development in places where subnational data are missing or not precise. Taken together, the research confirms four groups of important determinants of industrial growth: natural resources, agriculture growth, institutions, and manufacturing imports. Our findings indicate that Africa should follow a more multisector approach for development, putting natural resources and agriculture productivity growth at the forefront.
    Keywords: Industrial growth; Machine learning; Africa
    JEL: I32 O15 O40 O55
    Date: 2019–01
  2. By: van der Vorst, Tommy; Jelicic, Nick
    Abstract: In this study we explore the potential impact of educational AI applications in personalized learning. According to Bloom (1984) students that are tutored one-to-one perform two standard deviations better than students who learn via traditional educational methods. Due to the limited amount of teachers and costs associated, personalized one-to-one learning is not generally feasible from a societal point of view. Breakthroughs in the field of machine learning offer promising avenues to aid in personalized learning. AI may hence be the 'holy grail' in unlocking the potential of one-to-one learning, by enabling applications to offer personalized teaching to each individual student. We assess the potential impact of AI in personalized learning from a socio-technical perspective. Therefore, we investigate the technological possibilities, as well as any aspects that may impact adoption, e.g. legal, societal and ethical. To conclude we formulate policy options that can stimulate the adoption of AI-driven personalized learning applications.
    Date: 2019
  3. By: Junming Yang; Yaoqi Li; Xuanyu Chen; Jiahang Cao; Kangkang Jiang
    Abstract: Training a practical and effective model for stock selection has been a greatly concerned problem in the field of artificial intelligence. Even though some of the models from previous works have achieved good performance in the U.S. market by using low-frequency data and features, training a suitable model with high-frequency stock data is still a problem worth exploring. Based on the high-frequency price data of the past several days, we construct two separate models-Convolution Neural Network and Long Short-Term Memory-which can predict the expected return rate of stocks on the current day, and select the stocks with the highest expected yield at the opening to maximize the total return. In our CNN model, we propose improvements on the CNNpred model presented by E. Hoseinzade and S. Haratizadeh in their paper which deals with low-frequency features. Such improvements enable our CNN model to exploit the convolution layer's ability to extract high-level factors and avoid excessive loss of original information at the same time. Our LSTM model utilizes Recurrent Neural Network'advantages in handling time series data. Despite considerable transaction fees due to the daily changes of our stock position, annualized net rate of return is 62.27% for our CNN model, and 50.31% for our LSTM model.
    Date: 2019–11
  4. By: Marco Guerzoni; Consuelo R. Nava; Massimiliano Nuccio
    Abstract: This paper shows how data science can contribute to improving empirical research in economics by leveraging on large datasets and extracting information otherwise unsuitable for a traditional econometric approach. As a test-bed for our framework, machine learning algorithms allow us to create a new holistic measure of innovation built on a 2012 Italian Law aimed at boosting new high-tech firms. We adopt this measure to analyse the impact of innovativeness on a large population of Italian firms which entered the market at the beginning of the 2008 global crisis. The methodological contribution is organised in different steps. First, we train seven supervised learning algorithms to recognise innovative firms on 2013 firmographics data and select a combination of those with best predicting power. Second, we apply the former on the 2008 dataset and predict which firms would have been labelled as innovative according to the definition of the law. Finally, we adopt this new indicator as regressor in a survival model to explain firms' ability to remain in the market after 2008. Results suggest that the group of innovative firms are more likely to survive than the rest of the sample, but the survival premium is likely to depend on location.
    Date: 2019–11
  5. By: Nyström, Anna-Greta; Gugenishvili, Ilia
    Abstract: The question of how actors perceive business opportunities has puzzled both researchers and practitioners for decades. In the era of artificial intelligence, machine learning, and the Internet of things, many actors of the technology-intensive industries question how to use new technology to create value, and how to monetize new service concepts. In this paper, we focus on the next mobile communications technology, 5G, as one of the potential value-creators for the future that holds business opportunities for its utilizers and deployers. The concept of business opportunities is strongly associated with research on entrepreneurship (cf. Carlsson et al., 2003). Entrepreneurial opportunities consist of a set of ideas, beliefs, and actions that enable the introduction of goods, services, raw materials, and organizing methods in the absence of current markets for them (Sarasvathy et al., 2003). The research stream of entrepreneurial opportunities (cf. Alvarez & Barney, 2007, 2010; Dimov, 2007, 2011; Eckhardt & Shane, 2003) can offer new insights into the development of opportunities in high technologyintensive fields, and especially as regards the development of 5G. Strategies for opportunity identification, exploitation, and value creation are vital in the 5G era, as non-ICT traditional business sectors begin to deploy wireless technologies (e.g., factories, automotive, etc.). Researchers expect that 5G will change the business models and business ecosystems; it will also better address the evolving needs of customers (cf. Kliks et al., 2018). Unlike already existing mobile communications systems, 5G allows integration of vertical industries, e.g., energy, media, health, factories, and the automotive industry (5G-PPP, 2016). Thus, specialized companies will be able to provide services and establish positions in the value chains and actor networks. This is a major transformation compared to an environment dominated by bilateral relationships between mobile operators and their customers...
    Date: 2019
  6. By: Hector G. Lopez-Ruiz; Nora Nezamuddin; Reema Al Hassan; Abdelrahman Muhsen (King Abdullah Petroleum Studies and Research Center)
    Abstract: This paper focuses on the methodology for estimating total freight transport activity (FTA) for three countries — China, India and Saudi Arabia — with the objective of building on current state-of-the-art transportation modeling in three key areas: Studying the relationship between nighttime lights (NTL) and FTA allows for an estimation of full transportation datasets for countries where only a few observation points exist or where data is unavailable. Establishing the foundation for future work on how to use this approach in transport flow estimation (origin-destination matrices). Determining whether this approach can be used globally, given the coverage of the satellite data used. The paper uses the KAPSARC Transport Analysis Framework (KTAF), which estimates transport activity from freely available global data sources, satellite images and NTL. It is a tool for estimating freight transport activity that can be used in models to measure the impact of an accelerated transport policy planning approach. The methodology offers a solution to inadequate data access and allows for scenario building in policy planning for transportation. This approach allows for quick estimation of the effects of policy measures and economic changes on transportation activities at a global level. The paper also includes a detailed guide on how to replicate the methodology used in this analysis.
    Keywords: Freight Modeling, KAPSARC Transport Analysis Framework (KTAF), Nighttime lights satelite data, Transportation
    Date: 2019–05–09
  7. By: Philippe G. LeFloch; Jean-Marc Mercier
    Abstract: We review a numerical technique, referred to as the Transport-based Meshfree Method (TMM), and we discuss its applications to mathematical finance. We recently introduced this method from a numerical standpoint and investigated the accuracy of integration formulas based on the Monte-Carlo methodology: quantitative error bounds were discussed and, in this short note, we outline the main ideas of our approach. The techniques of transportation and reproducing kernels lead us to a very efficient methodology for numerical simulations in many practical applications, and provide some light on the methods used by the artificial intelligence community. For applications in the finance industry, our method allows us to compute many types of risk measures with an accurate and fast algorithm. We propose theoretical arguments as well as extensive numerical tests in order to justify sharp convergence rates, leading to rather optimal computational times. Cases of direct interest in finance support our claims and the importance of the problem of the curse of dimensionality in finance applications is briefly discussed.
    Date: 2019–11
  8. By: Garcia-Murillo, Martha; MacInnes, Ian
    Abstract: Artificial Intelligence (AI) is likely to have a significant impact on work. Examples from the past demonstrate that it has created jobs but also displaced workers. The primary question this study aims to answer is what have been the effects that previous revolutionary computing technologies have had and how have institutional values shaped the way workers were affected. The paper involves a historical analysis of the experiences that society in the United States has had with technological innovation. The research relies on academic, government, and trade publications of earlier periods in the development of computer technology. In this effort, we examine the literature on institutional economics to help us understand the way society has transitioned and the forces that have shaped the outcomes. Institutional economics has two main branches that explain change: the ceremonial and the instrumental. The ceremonial values perspective focuses on the customs and conventions that prevail in a community. The instrumental perspective focuses on a society's processes of inquiry, acquisition of knowledge, and use of scientific inquiry to solve problems Our analysis suggests that in all of these periods initial implementations suffered from installation problems, system bugs, and troubleshooting frustrations that generated employment; however, as the technology improves, it is likely to enhance productivity, but displace, workers. Up to this point, the U.S. government has not been able to respond adequately to the challenge. We attribute this to the ceremonial values that public officials and society entertain about personal responsibility and small government.
    Keywords: Artificial intelligence (AI),Technological displacement,Economic transition,Ceremonial values,Instrumental values,Public policy
    Date: 2019
  9. By: Magnus Wiese; Lianjun Bai; Ben Wood; Hans Buehler
    Abstract: We construct realistic equity option market simulators based on generative adversarial networks (GANs). We consider recurrent and temporal convolutional architectures, and assess the impact of state compression. Option market simulators are highly relevant because they allow us to extend the limited real-world data sets available for the training and evaluation of option trading strategies. We show that network-based generators outperform classical methods on a range of benchmark metrics, and adversarial training achieves the best performance. Our work demonstrates for the first time that GANs can be successfully applied to the task of generating multivariate financial time series.
    Date: 2019–11
  10. By: Longstaff, P.H.
    Abstract: This paper sets out the current discussions around the world on the issue of Artificial Intelligence. Several themes seem to be apparent and are set out here. It is important to know that this is a snapshot of these discussions and the author apologizes in advance that research has only been done in English. The top few topics being discussed are then briefly discussed as areas where all governments may put some of their efforts.
    Date: 2019
  11. By: Victor Chernozhukov; Jerry A. Hausman; Whitney K. Newey
    Abstract: From its inception, demand estimation has faced the problem of "many prices." This paper provides estimators of average demand and associated bounds on exact consumer surplus when there are many prices in cross-section or panel data. For cross-section data we provide a debiased machine learner of consumer surplus bounds that allows for general heterogeneity and solves the "zeros problem" of demand. For panel data we provide bias corrected, ridge regularized estimators of average coefficients and consumer surplus bounds. In scanner data we find smaller panel elasticities than cross-section and that soda price increases are regressive.
    JEL: C13 C14 C21 C23 C55 D12
    Date: 2019–11

This nep-big issue is ©2019 by Tom Coupé. 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.