nep-cmp New Economics Papers
on Computational Economics
Issue of 2021‒01‒11
24 papers chosen by
Stan Miles
Thompson Rivers University

  1. Solving the joint order batching and picker routing problem, as a clustered vehicle routing problem By AERTS, Babiche; CORNELISSENS, Trijntje; SÖRENSEN, Kenneth
  2. Drivers of automation and consequences for jobs in engineering services: an agent-based modelling approach By Kyvik Nordås, Hildegunn; Klügl, Franziska
  3. QMM A Quarterly Macroeconomic Model of the Icelandic Economy Version 4.0 By à sgeir Daníelsson; Lúdvik Elíasson; Magnús F. Gudmundsson; Svava J. Haraldsdóttir; Lilja S. Kro; Thórarinn G. Pétursson; Thorsteinn S. Sveinsson
  4. Exploration of model performances in the presence of heterogeneous preferences and random effects utilities awareness By Nikita Gusarov; Amirreza Talebijamalabad; Iragaël Joly
  5. A machine learning solver for high-dimensional integrals: Solving Kolmogorov PDEs by stochastic weighted minimization and stochastic gradient descent through a high-order weak approximation scheme of SDEs with Malliavin weights By Riu Naito; Toshihiro Yamada
  6. Инфраструктура и экономический рост. «Бюджетный маневр» в России By Dmitriy, Skrypnik
  7. Machine Learning Advances for Time Series Forecasting By Ricardo P. Masini; Marcelo C. Medeiros; Eduardo F. Mendes
  8. Trader-Company Method: A Metaheuristic for Interpretable Stock Price Prediction By Katsuya Ito; Kentaro Minami; Kentaro Imajo; Kei Nakagawa
  9. Information Entropy, Continuous Improvement, and US Energy Performance: A Novel Stochastic-Entropic Analysis for Ideal Solutions (SEA-IS) By Jorge Antunes; Rangan Gupta; Zinnia Mukherjee; Peter Wanke
  10. Fiscal DSGE Model for Latvia By Ginters Buss; Patrick Gruning
  11. Endogenous Long-Term Productivity Performance in Advanced Countries: A Novel Two-Dimensional Fuzzy-Monte Carlo Approach By Jorge Antunes; Goodness C. Aye; Rangan Gupta; Peter Wanke
  12. Forecasting Mid-price Movement of Bitcoin Futures Using Machine Learning By Akyildirim, Erdinc; Cepni, Oguzhan; Corbet, Shaen; Uddin, Gazi Salah
  13. Fraud detection by a multinomial model: Separating honesty from unobserved fraud By Andersson, Jonas; Olden, Andreas; Rusina, Aija
  14. The Value Added of Machine Learning to Causal Inference: Evidence from Revisited Studies By Anna Baiardi; Andrea A. Naghi
  15. A Comparison of Statistical and Machine Learning Algorithms for Predicting Rents in the San Francisco Bay Area By Paul Waddell; Arezoo Besharati-Zadeh
  16. DYNIMO - Version III. A DSGE model of the Icelandic economy By Stefán Thórarinsson
  17. ABC: An Agent Based Exploration of the Macroeconomic Effects of Covid-19 By Domenico Delli Gatti; Severin Reissl
  18. Locus of control and Human Capital Investment Decisions: The Role of Effort, Parental Preferences and Financial Constraints By Szabó-Morvai Ágnes; Hubert János Kiss
  19. Recent Trends in Real Estate Research: A Comparison of Recent Working Papers and Publications using Machine Learning Algorithms By Breuer, Wolfgang; Steininger, Bertram
  20. Clusters in UK Self-Employment By Jack Blundell
  21. The Missing 15 Percent of Patent Citations By Cyril Verluise; Gabriele Cristelli; Kyle Higham; Gaetan de Rassenfosse
  22. First Time Around: Local Conditions and Multi-dimensional Integration of Refugees By Cevat Giray Aksoy; Panu Poutvaara; Felicitas Schikora
  23. The Cross-Sectional Pricing of Corporate Bonds Using Big Data and Machine Learning By Turan G. Bali; Amit Goyal; Dashan Huang; Fuwei Jiang; Quan Wen
  24. The Unintended Consequences of Stay-at-Home Policies on Work Outcomes: The Impacts of Lockdown Orders on Content Creation By Xunyi Wang; Reza Mousavi; Yili Hong

  1. By: AERTS, Babiche; CORNELISSENS, Trijntje; SÖRENSEN, Kenneth
    Abstract: ?The Joint Order Batching and Picker Routing Problem (JOBPRP) is a very promising approach to minimize the order picking travel distance in a picker-to-parts warehouse environment. We show that this JOBPRP can be modelled as a clustered vehicle routing problem (CluVRP) by replacing vehicles by batches, clusters by orders and customers by pick locations. To solve this cluster-based model of the JOBPRP,we implement a two-level Variable Neighborhood Search (2level-VNS) meta-heuristic as used earlier for the CluVRP, and study which adaptations are required to perform e ciently in a warehouse environment. Additionally, we test if the Hausdor? distance used for the CluVRP can serve as a valid clustering criterion for order batching. We implement the Hausdor? distance in two di?erent ways in our batching heuristic, and compare the performance with the cumulative minimal aisles visited-criterion, known as a well-performing batching metric in rectangular warehouses with parallel aisles. Finally, we show that the CluVRP model solved by the 2level-VNS approach performs well compared to state-of-the-art algorithms for the OBP in single-block warehouses. Only a multi-start VNS approach published recently obtains slightly be?er solutions. Concerning the Hausdor? distance, we must conclude that in most experiments the minimum-aisles criterion retains a be?er ?t in this warehouse context.
    Date: 2020–04
  2. By: Kyvik Nordås, Hildegunn (Örebro University School of Business); Klügl, Franziska (School of Science and Technology)
    Abstract: This paper studies the uptake of AI-driven automation and its impact on employment, using a dynamic agent-based model (ABM). It simulates the adoption of automation software as well as job destruction and job creation in its wake. There are two types of agents: manufacturing firms and engineering services firms. The agents choose between two business models: consulting or automated software. From the engineering firms’ point of view, the model exhibits static economies of scale in the software model and dynamic (learning by doing) economies of scale in the consultancy model. From the manufacturing firms’ point of view, switching to the software model requires restructuring of production and there are network effects in switching. The ABM matches engineering and manufacturing agents and derives employment of engineers and the tasks they perform, i.e. consultancy, software development, software maintenance, or employment in manufacturing. Policy parameters influencing the results are occupational licensing and protection of intellectual property rights. We find that the uptake of software is gradual; slow in the first few years and then accelerates. Software is fully adopted after about 18 years in the base line run. The adoption rate is slower the higher the license fee for software, while the adoption rate is faster the higher the mark-up rate of consultancy. Employment of engineers shifts from consultancy to software development and to new jobs in manufacturing. Spells of unemployment may occur, if skilled jobs creation in manufacturing is slow. Finally, the model generates boom and bust cycles in the software sector.
    Keywords: Technology Uptake; Employment; Automation; Economic Modelling; Agent-Based Simulation
    JEL: C51 C61 J44 L84 O33
    Date: 2020–12–23
  3. By: Ã sgeir Daníelsson; Lúdvik Elíasson; Magnús F. Gudmundsson; Svava J. Haraldsdóttir; Lilja S. Kro; Thórarinn G. Pétursson; Thorsteinn S. Sveinsson
    Abstract: This Handbook documents Version 4.0 of the Quarterly Macroeconomic Model of the Central Bank of Iceland (QMM). QMM and the underlying quarterly database have been under construction since 2001 at the Research and Forecasting Division of the Economics and Monetary Policy Department at the Bank and was first implemented in the forecasting round for the Monetary Bulletin 2006/1 in March 2006. QMM is used by the Bank for forecasting and various policy simulations and therefore plays a key role as an organisational framework for viewing the medium-term future when formulating monetary policy at the Bank. This paper is mainly focused on the short and mediumterm properties of QMM. Steady state properties of the model are documented in a paper by Dan’elsson (2009).
    Date: 2019–12
  4. By: Nikita Gusarov (GAEL - Laboratoire d'Economie Appliquée de Grenoble - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement - Grenoble INP - Institut polytechnique de Grenoble - Grenoble Institute of Technology - UGA - Université Grenoble Alpes - CNRS - Centre National de la Recherche Scientifique - UGA - Université Grenoble Alpes); Amirreza Talebijamalabad (Grenoble INP - Institut polytechnique de Grenoble - Grenoble Institute of Technology - UGA - Université Grenoble Alpes); Iragaël Joly (GAEL - Laboratoire d'Economie Appliquée de Grenoble - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement - Grenoble INP - Institut polytechnique de Grenoble - Grenoble Institute of Technology - UGA - Université Grenoble Alpes - CNRS - Centre National de la Recherche Scientifique - UGA - Université Grenoble Alpes)
    Abstract: This work is a cross-disciplinary study of econometrics and machine learning (ML) models applied to consumer choice preference modelling. To bridge the interdisciplinary gap, a simulation and theorytesting framework is proposed. It incorporates all essential steps from hypothetical setting generation to the comparison of various performance metrics. The flexibility of the framework in theory-testing and models comparison over economics and statistical indicators is illustrated based on the work of Michaud, Llerena and Joly (2012). Two datasets are generated using the predefined utility functions simulating the presence of homogeneous and heterogeneous individual preferences for alternatives' attributes. Then, three models issued from econometrics and ML disciplines are estimated and compared. The study demonstrates the proposed methodological approach's efficiency, successfully capturing the differences between the models issued from different fields given the homogeneous or heterogeneous consumer preferences.
    Keywords: Discrete choice models,Neural network analysis,Performance comparison,Heterogeneous preferences
    Date: 2020–10
  5. By: Riu Naito; Toshihiro Yamada
    Abstract: The paper introduces a very simple and fast computation method for high-dimensional integrals to solve high-dimensional Kolmogorov partial differential equations (PDEs). The new machine learning-based method is obtained by solving a stochastic weighted minimization with stochastic gradient descent which is inspired by a high-order weak approximation scheme for stochastic differential equations (SDEs) with Malliavin weights. Then solutions to high-dimensional Kolmogorov PDEs or expectations of functionals of solutions to high-dimensional SDEs are accurately approximated without suffering from the curse of dimensionality. Numerical examples for PDEs and SDEs up to 100 dimensions are shown by using second and third-order discretization schemes in order to demonstrate the effectiveness of our method.
    Date: 2020–12
  6. By: Dmitriy, Skrypnik
    Abstract: The paper examines the world experience in stimulating economic development based on the creation of infrastructure. The effects on economic growth are found to be moderate. Building infrastructure boosts economic growth in the poorest countries, where infrastructure shortages are critical. For developing countries, the creation of infrastructure is of secondary importance, and infrastructure has a significant impact when it is part of development projects implemented by industrial policy measures based on a system of catch-up development institutions. For developed countries, infrastructure can lag behind market sector dynamics and investment can again have significant impact. For the Russian economy, where the stock and quality of infrastructural and human capital are at medium levels, it is not necessary to expect accelerated growth from the implementation of national projects, the bulk of which goes to the creation of infrastructure. This conclusion is confirmed by numerical experiments based on a computable general economic equilibrium model constructed in this work for Russia. At the same time, it is found that the increase in the VAT rate itself, which is part of the budget maneuver, leads to a decrease in output in most industries, including the manufacturing sector. In the scenario of a budget maneuver - with an increase in the VAT rate and an increase in government spending - the negative effects of the increase in VAT are intensified: there is stronger growth in the public sector, construction and raw materials, which ensures economic growth, but at the same time increases the cost of factors and leads to a deepening decline in other sectors as a result, the economy experiences a double negative impact. The public sector begins to reproduce the mechanism of the Dutch disease.
    Keywords: infrastructure, government spending, fiscal policy, economic growth, computable general equilibrium models, structural vector autoregression models
    JEL: C68 E16 E17 E62 H25 H50 H54 O23 O43
    Date: 2020–11–15
  7. By: Ricardo P. Masini; Marcelo C. Medeiros; Eduardo F. Mendes
    Abstract: In this paper we survey the most recent advances in supervised machine learning and high-dimensional models for time series forecasting. We consider both linear and nonlinear alternatives. Among the linear methods we pay special attention to penalized regressions and ensemble of models. The nonlinear methods considered in the paper include shallow and deep neural networks, in their feed-forward and recurrent versions, and tree-based methods, such as random forests and boosted trees. We also consider ensemble and hybrid models by combining ingredients from different alternatives. Tests for superior predictive ability are briefly reviewed. Finally, we discuss application of machine learning in economics and finance and provide an illustration with high-frequency financial data.
    Date: 2020–12
  8. By: Katsuya Ito; Kentaro Minami; Kentaro Imajo; Kei Nakagawa
    Abstract: Investors try to predict returns of financial assets to make successful investment. Many quantitative analysts have used machine learning-based methods to find unknown profitable market rules from large amounts of market data. However, there are several challenges in financial markets hindering practical applications of machine learning-based models. First, in financial markets, there is no single model that can consistently make accurate prediction because traders in markets quickly adapt to newly available information. Instead, there are a number of ephemeral and partially correct models called "alpha factors". Second, since financial markets are highly uncertain, ensuring interpretability of prediction models is quite important to make reliable trading strategies. To overcome these challenges, we propose the Trader-Company method, a novel evolutionary model that mimics the roles of a financial institute and traders belonging to it. Our method predicts future stock returns by aggregating suggestions from multiple weak learners called Traders. A Trader holds a collection of simple mathematical formulae, each of which represents a candidate of an alpha factor and would be interpretable for real-world investors. The aggregation algorithm, called a Company, maintains multiple Traders. By randomly generating new Traders and retraining them, Companies can efficiently find financially meaningful formulae whilst avoiding overfitting to a transient state of the market. We show the effectiveness of our method by conducting experiments on real market data.
    Date: 2020–12
  9. By: Jorge Antunes (COPPEAD Graduate Business School, Federal University of Rio de Janeiro, Rua Paschoal, Lemme, 355, 21949-900 Rio de Janeiro, Brazil); Rangan Gupta (Department of Economics, University of Pretoria, Pretoria, 0002, South Africa); Zinnia Mukherjee (Department of Economics, Simmons University, 300 The Fenway, Boston, MA 02115, USA); Peter Wanke (COPPEAD Graduate Business School, Federal University of Rio de Janeiro, Rua Paschoal, Lemme, 355, 21949-900 Rio de Janeiro, Brazil)
    Abstract: Previous energy performance studies neglected the role of information entropy in feedback processes between input and output slacks. Superior energy performance may be achieved through the capability of learning from how increased outputs could yield reduced inputs and vice-versa. This paper focus on this gap, by presenting an assessment of US states for a 35-year period in lieu of relevant socio-economic and demographic variables. US is the world largest energy producer and consumer, being well-known not only for innovation in efficient energy use but also for managerial feedback mechanisms in the energy field which ensures continuous improvement in generation and consumption. First, a novel SEA-IS (Stochastic-Entropic Analysis for Ideal Solutions) model is developed to assess the potential information gains that may arise from energy slacks minimization given different optimal reduction quantiles in US states. This non-linear stochastic optimization model not only relies on Beta distributed priors to model the odds-ratio of learning feedback but also takes advantages of numerous strengths present in DEA and TOPSIS approaches for performance management. Machine learning methods are also employed to predict information gains in terms of contextual variables. Results indicate that California is the only U.S. state that has indicate strong mutual information feedback and continuous improvements in efficiency. There is ample scope for harnessing the power of information gains in improving energy efficiency, particularly in 37 U.S. states, which indicates scope for a public-private partnership to achieve this goal.
    Keywords: US energy, performance, state-level, stochastic-entropic approach, information gains, slack management, feedback
    Date: 2020–12
  10. By: Ginters Buss (Latvijas Banka); Patrick Gruning (CEFER, Lietuvos Bankas)
    Abstract: We develop a fiscal dynamic stochastic general equilibrium (DSGE) model for policy simulation and scenario analysis purposes tailored to Latvia, a small open economy in a monetary union. The fiscal sector elements comprise government investment, government consumption, government transfers that are asymmetrically directed to both optimizing and hand-to-mouth households, cyclical unemployment benefits, foreign ownership of government debt, import content in public consumption and investment, and fiscal rules for each fiscal instrument. The model features a search-and-matching labour market friction with pro-cyclical labour costs, a financial accelerator mechanism, and import content in final goods. We estimate the model using Latvian data, study the new channels in the model, and provide a comprehensive analysis on the macroeconomic effects of the fiscal elements. A particular finding is that having foreign ownership of government debt generally breaks the Ricardian equivalence paradigm.
    Keywords: small open economy, fiscal policy, fiscal rules, Bayesian estimation
    JEL: E0 E2 E3 F4 H2 H3 H6
    Date: 2020–12–15
  11. By: Jorge Antunes (COPPEAD Graduate Business School, Federal University of Rio de Janeiro, Rua Paschoal, Lemme, 355, 21949-900 Rio de Janeiro, Brazil); Goodness C. Aye (Department of Economics, University of Pretoria, Pretoria, 0002, South Africa); Rangan Gupta (Department of Economics, University of Pretoria, Pretoria, 0002, South Africa); Peter Wanke (COPPEAD Graduate Business School, Federal University of Rio de Janeiro, Rua Paschoal, Lemme, 355, 21949-900 Rio de Janeiro, Brazil)
    Abstract: Long-term productivity performance at the country level has been a research object under different theoretical lenses, scrutinized by different modelling approaches. This paper revisits the dataset used in Bergeaud et al. (2016) and investigates the endogenous sources of distinct productivity performance in different advanced countries. Country-level data series with more than one hundred year time-span were collected for each one of the following attributes: Labor productivity (LP), Total Factor Productivity (TFP), Capital Intensity (KI), Gross Domestic Product per capita (GDP pc), Average age of equipment capital stock (Age K), and Human Capital intensity (Human K).Differently from previous studies, a Two-Dimensional Fuzzy-Monte Carlo Analysis (2DFMC) approach is proposed here to decompose the sources of long-term productivity performance. In the first dimension, a novel multi-attribute decision-making (MADM) model based on Type-2 Fuzzy Sets (T2FS) is developed to compute and rank long-term productivity performance of each county using Unbiased-Power functions for Ideal Solutions (UP-IS). Next, in the second dimension, a Stochastic Structural Relationship Programming (SSRP) Model based on neural networks is proposed to evaluate the endogenous feedbacks among the aforementioned productivity attributes and overall productivity performance. Results suggest that the UP-IS presented higher cross-performance scores relative to the TOPSIS base-case. Norway is the best performing country with a positive performance score of 0.854 while Portugal is the worst with a score of 0.347. In terms of ordinal ranking of long-term productivity performance, UP-IS ranked first, followed by TPF and next LP and GDPpc. Further, Human K and Age K have positive and negative impact respectively on long-term productivity performance in advanced countries. On the other hand, productivity performance has positive impact on KI and TFP but negative impact on LP and GDPpc.
    Keywords: endogeneity, type-2 fuzzy sets, 2DFMC, stochastic performance, long-term productivity, advanced countries
    Date: 2020–12
  12. By: Akyildirim, Erdinc (Department of Mathematics, ETH, Zurich, Switzerland and University of Zurich, Department of Banking and Finance, Zurich, Switzerland and Department of Banking and Finance, Burdur Mehmet Akif Ersoy University, Burdur, Turkey); Cepni, Oguzhan (Department of Economics, Copenhagen Business School); Corbet, Shaen (DCU Business School, Dublin City University, Dublin 9, Ireland and School of Accounting, Finance and Economics, University of Waikato, New Zealand); Uddin, Gazi Salah (Department of Management and Engineering, Linköping University, 581 83 Linköping, Sweden)
    Abstract: In the aftermath of the global financial crisis and on-going COVID-19, investors face challenges in understanding price dynamics across assets. In this paper, we explore the applicability of a large scale comparison of machine learning algorithms (MLA) to predict mid-price movement for bitcoin futures prices. We use high-frequency intra-day data to evaluate the relative forecasting performances across various time-frequencies, ranging between 5-minutes and 60-minutes. The empirical analysis is based on six different specifications of MLA methods during periods of pandemic. The empirical results show that MLA outperforms the random walk and ARIMA forecasts in Bitcoin futures markets, which may have important implications in the decision-making process of predictability.
    Keywords: Cryptocurrency; Bitcoin futures; Machine learning; Covid-19; k-Nearest neighbors; Logistic regression; Naive bayes; Random forest; Support vector machine; Extreme gradient; Boosting
    JEL: C60 E50
    Date: 2020–12–22
  13. By: Andersson, Jonas (Dept. of Business and Management Science, Norwegian School of Economics); Olden, Andreas (Dept. of Business and Management Science, Norwegian School of Economics); Rusina, Aija (Dept. of Business and Management Science, Norwegian School of Economics)
    Abstract: In this paper we investigate the EM-estimator of the model by Caudill et al. (2005). The purpose of the model is to identify items, e.g. individuals or companies, that are wrongly classified as honest; an example of this is the detection of tax evasion. Normally, we observe two groups of items, labeled fraudulent and honest, but suspect that many of the observationally honest items are, in fact, fraudulent. The items observed as honest are therefore divided into two unobserved groups, honestH, representing the truly honest, and honestF, representing the items that are observed as honest, but that are actually fraudulent. By using a multinomial logit model and assuming commonality between the observed fraudulent and the unobserved honestF, Caudill et al. (2005) present a method that uses the EM-algorithm to separate them. By means of a Monte Carlo study, we investigate how well the method performs, and under what circumstances. We also study how well bootstrapped standard errors estimates the standard deviation of the parameter estimators.
    Keywords: Fraud detection; EM-algorithm; multinomial logit model; Monte Carlo study
    JEL: C00 C10
    Date: 2020–12–31
  14. By: Anna Baiardi (Erasmus University Rotterdam); Andrea A. Naghi (Erasmus University Rotterdam)
    Abstract: A new and rapidly growing econometric literature is making advances in the problem of using machine learning (ML) methods for causal inference questions. Yet, the empirical economics literature has not started to fully exploit the strengths of these modern methods. We revisit influential empirical studies with causal machine learning methods and identify several advantages of using these techniques. We show that these advantages and their implications are empirically relevant and that the use of these methods can improve the credibility of causal analysis.
    Keywords: Machine learning, causal inference, average treatment effects, heterogeneous treatment effects
    JEL: D04 C01 C21
    Date: 2021–01–04
  15. By: Paul Waddell; Arezoo Besharati-Zadeh
    Abstract: Urban transportation and land use models have used theory and statistical modeling methods to develop model systems that are useful in planning applications. Machine learning methods have been considered too 'black box', lacking interpretability, and their use has been limited within the land use and transportation modeling literature. We present a use case in which predictive accuracy is of primary importance, and compare the use of random forest regression to multiple regression using ordinary least squares, to predict rents per square foot in the San Francisco Bay Area using a large volume of rental listings scraped from the Craigslist website. We find that we are able to obtain useful predictions from both models using almost exclusively local accessibility variables, though the predictive accuracy of the random forest model is substantially higher.
    Date: 2020–11
  16. By: Stefán Thórarinsson
    Abstract: This handbook describes the third version of DYNIMO, the Central Bank of Iceland’s dynamic stochastic general equilibrium model. Derivations from first principles to final equations are presented. In addition, the more advanced mathematical tools needed for the derivations are stated and their validity motivated. Subsequently, we estimate the model on Icelandic data over the period 2011Q1-2019Q4. Finally, evaluation of the model’s properties is performed.
    JEL: C32 C51 F41
    Date: 2020–12
  17. By: Domenico Delli Gatti; Severin Reissl
    Abstract: We employ a new macro-epidemiological agent based model to evaluate the “lives vs livelihoods” trade-off brought to the fore by Covid-19. The disease spreads across the networks of agents’ social and economic contacts and feeds back on the economic dimension of the model through various channels such as employment and consumption demand. We show that under a lockdown scenario the model is able to closely reproduce the epidemiological dynamics of the first wave of the coronavirus epidemic in Lombardy. We then explore the efficacy of the fiscal response to Covid-19 which may take different routes: income support, liquidity provision, credit guarantees. In an agent based setting we gain additional insights on the way in which fiscal measures impact not only on GDP but also on the defaults of firms and the allocation of inputs. We find that liquidity support for firms, a short-time working scheme with compensation for workers, and direct transfer payments to households are effective policy tools to alleviate the economic impact of the epidemic and the lockdown.
    Keywords: agent-based models, epidemic, Covid, fiscal policy
    JEL: E21 E22 E24 E27 E62 E65
    Date: 2020
  18. By: Szabó-Morvai Ágnes (KRTK-KTI, 1097 Budapest, Tóth Kálmán u. 4.és Debreceni Egyetem, 4032 Debrecen, Böszörményi út 138.); Hubert János Kiss (KRTK KTI, 1097 Budapest, Tóth Kálmán u. 4, Magyarország. andBudapesti Corvinus Egyetem, 1093 Budapest, Fõvám tér 8, Magyarország.)
    Abstract: We study the relationship between locus of control (LoC) and human capital investment decisions in the adolescence, using PDS lasso to exploit high-dimensional data. While LoC is not significantly associated with graduation from high school once we use exogenous controls, it correlates strongly with dropout age and college attendance even if we take into account predetermined variables and cognitive abilities, and it exhibits a significant positive relationship with plans to apply to college even if we control for potentially endogenous variables. We find that effort is an important conduit through which LoC operates and it is different from the expectation channel that has been already documented in the literature. The associations are heterogenous: LoC has a significant association with dropout age, high school graduation and college application plans in low-SES families, and with college attendance in mid-SES families. These heterogenous relations are in a large part determined by parental preferences and financial constraints.
    Keywords: Human Capital Investment Decision, LoC, Machine learning, PDS Lasso
    JEL: D91 I21 I23 I24 I26
    Date: 2020–12
  19. By: Breuer, Wolfgang (RWTH Aachen University, Department of Finance, Aachen, Germany); Steininger, Bertram (Department of Real Estate and Construction Management, Royal Institute of Technology)
    Abstract: This paper is organized as follows. In Section 1, we describe the economic relevance of the real estate sector and its recent dynamics. Then, we identify the most mentioned keywords of working papers presented at the real estate conferences between 2015 and 2019 and showing network figures for them in Section 2. In order to identify the newest trends, we rely on working papers since they have an average lead time of at least 1 to 2 years before they are published. In addition, we give a short overview of the articles published in this special issue. To get a better overview of the relevance of real estate related topics in finance, we analyzed the most relevant finance conferences and journals between 2015 and May 2020 in Section 3. To find the topics, we apply the text mining approach Latent Dirichlet Allocation (LDA), an unsupervised machine learning method. The real estate trends (retail, e-commerce) and the potential impact of COVID-19 is described in Section 4.
    Keywords: Recent trends; Real estate; Machine learning; Latent Dirichlet Allocation; LDA
    JEL: C45 C80 R30
    Date: 2020–12–28
  20. By: Jack Blundell
    Abstract: UK Self-employment has soared in recent years. With existing labour market policy designed to cater for conventional employee relationships, policymakers in this field are increasingly seeking to better understand these workers' characteristics in order to ensure that new labour market regulations are designed appropriately, and are targeted towards the groups that require social assistance. In this paper I apply a machine learning method to ask whether there exist distinct 'clusters' of workers within self-employment, corresponding to groups with similar observable characteristics. My analysis first uncovers a two-group typology, with a distinct divide between a low-educated male group and a high-educated female group. While groups differ on characteristics, drawing on new survey data I find that both are similarly satisfied with self-employment. I also uncover a six-group typology. This detailed clustering reveals a sub-group of low-educated young men who are dissatisfied with self-employment and are most likely to report self-employment as their only employment option, many of whom can be broadly classified as 'gig economy' workers.
    Keywords: self-employment,gig economy, worker characteristics
    Date: 2020–02
  21. By: Cyril Verluise (Collège de France); Gabriele Cristelli (Ecole polytechnique federale de Lausanne); Kyle Higham (Hitotsubashi University); Gaetan de Rassenfosse (Ecole polytechnique federale de Lausanne)
    Abstract: Patent citations are one of the most commonly-used metrics in the innovation literature. Leading uses of patent-to-patent citations are associated with the quantification of inventions’ quality and the measurement of knowledge flows. Due to their widespread availability, scholars have exploited citations listed on the front-page of patent documents. Citations appearing in the full-text of patent documents have been neglected. We apply modern machine learning methods to extract these citations from the text of USPTO patent documents. Overall, we are able to recover an additional 15 percent of patent citations that could not be found using only front-page data. We show that in-text citations bring a different type of information compared to front-page citations. They exhibit higher text-similarity to the citing patents and alter the ranking of patent importance. The dataset is available at (CC-BY-4).
    Keywords: Citation; Patent; Open data
    JEL: C81 O30
    Date: 2020–12
  22. By: Cevat Giray Aksoy; Panu Poutvaara; Felicitas Schikora
    Abstract: We study the causal effect of local labor market conditions and attitudes towards immigrants at the time of arrival on refugees’ multi-dimensional integration outcomes (economic, linguistic, navigational, political, psychological, and social). Using a unique dataset on refugees, we leverage a centralized allocation policy in Germany where refugees were exogenously assigned to live in specific counties. We find that high initial local unemployment negatively affects refugees’ economic and social integration: they are less likely to be in education or employment and they earn less. We also show that favorable attitudes towards immigrants promote refugees’ economic and social integration. The results suggest that attitudes toward immigrants are as important as local unemployment rates in shaping refugees’ integration outcomes. Using a machine learning classifier algorithm, we find that our results are driven by older people and those with secondary or tertiary education. Our findings highlight the importance of both initial economic and social conditions for facilitating refugee integration, and have implications for the design of centralized allocation policies.
    Keywords: International migration, refugees, integration, allocation policy
    JEL: F22 J15 J24
    Date: 2020
  23. By: Turan G. Bali (Georgetown University - Robert Emmett McDonough School of Business); Amit Goyal (University of Lausanne; Swiss Finance Institute); Dashan Huang (Singapore Management University - Lee Kong Chian School of Business); Fuwei Jiang (Central University of Finance and Economics (CUFE)); Quan Wen (Georgetown University - Department of Finance)
    Abstract: We provide a comprehensive study on the cross-sectional predictability of corporate bond returns using big data and machine learning. We examine whether a large set of equity and bond characteristics drive the expected returns on corporate bonds. Using either set of characteristics, we find that machine learning methods substantially improve the out-of-sample predictive power for bond returns, compared to the traditional linear regression models. While equity characteristics produce significant explanatory power for bond returns, their incremental predictive power relative to bond characteristics is economically and statistically insignificant. Bond characteristics provide as strong forecasting power for future equity returns as using equity characteristics alone. However, bond characteristics do not offer additional predictive power above and beyond equity characteristics when we combine both sets of predictors.
    Keywords: machine learning, big data, corporate bond returns, cross-sectional return predictability
    JEL: G10 G11 C13
    Date: 2020–09
  24. By: Xunyi Wang; Reza Mousavi; Yili Hong
    Abstract: The COVID-19 pandemic has posed an unprecedented challenge to individuals around the globe. To mitigate the spread of the virus, many states in the U.S. issued lockdown orders to urge their residents to stay at their homes, avoid get-togethers, and minimize physical interactions. While many offline workers are experiencing significant challenges performing their duties, digital technologies have provided ample tools for individuals to continue working and to maintain their productivity. Although using digital platforms to build resilience in remote work is effective, other aspects of remote work (beyond the continuation of work) should also be considered in gauging true resilience. In this study, we focus on content creators, and investigate how restrictions in individual's physical environment impact their online content creation behavior. Exploiting a natural experimental setting wherein four states issued state-wide lockdown orders on the same day whereas five states never issued a lockdown order, and using a unique dataset collected from a short video-sharing social media platform, we study the impact of lockdown orders on content creators' behaviors in terms of content volume, content novelty, and content optimism. We combined econometric methods (difference-in-differences estimations of a matched sample) with machine learning-based natural language processing to show that on average, compared to the users residing in non-lockdown states, the users residing in lockdown states create more content after the lockdown order enforcement. However, we find a decrease in the novelty level and optimism of the content generated by the latter group. Our findings have important contributions to the digital resilience literature and shed light on managers' decision-making process related to the adjustment of employees' work mode in the long run.
    Date: 2020–11

This nep-cmp issue is ©2021 by Stan Miles. 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.