nep-cmp New Economics Papers
on Computational Economics
Issue of 2021‒09‒27
thirteen papers chosen by
Stan Miles
Thompson Rivers University

  1. Deep Neural Network Algorithms for Parabolic PIDEs and Applications in Insurance Mathematics By R\"udiger Frey; Verena K\"ock
  2. Addressing Sample Selection Bias for Machine Learning Methods By Dylan Brewer; Alyssa Carlson
  3. Predicting Student Dropout: A Replication Study Based on Neural Networks By Jascha Buchhorn; Berthold U. Wigger
  4. Financial Trading with Feature Preprocessing and Recurrent Reinforcement Learning By Lin Li
  5. Catching the Drivers of Inclusive Growth in Sub-Saharan Africa: An Application of Machine Learning By Isaac K. Ofori
  6. Nowcasting aggregate services trade By Alexander Jaax; Frédéric Gonzales; Annabelle Mourougane
  7. Construction of Control Systems of Flow Parameters of the Smart Conveyor using a Neural Network By Pihnastyi, Oleh; Sytnikova, Anastasiya
  8. Fragile Algorithms and Fallible Decision-Makers: Lessons from the Justice System By Jens Ludwig; Sendhil Mullainathan
  9. Scenario generation for market risk models using generative neural networks By Solveig Flaig; Gero Junike
  10. Using Satellite Imagery and Machine Learning to Estimate the Livelihood Impact of Electricity Access By Nathan Ratledge; Gabriel Cadamuro; Brandon De la Cuesta; Matthieu Stigler; Marshall Burke
  11. Inheritances and Wealth Inequality: a Machine Learning Approach By Pedro Salas-Rojo; Juan Gabriel Rodríguez
  12. Human Resources in Europe. Estimation, Clusterization, Machine Learning and Prediction By Leogrande, Angelo; Costantiello, Alberto
  13. An AI-assisted Economic Model of Endogenous Mobility and Infectious Diseases: The Case of COVID-19 in the United States By Lin William Cong; Ke Tang; Bing Wang; Jingyuan Wang

  1. By: R\"udiger Frey; Verena K\"ock
    Abstract: In recent years a large literature on deep learning based methods for the numerical solution partial differential equations has emerged; results for integro-differential equations on the other hand are scarce. In this paper we study deep neural network algorithms for solving linear and semilinear parabolic partial integro-differential equations with boundary conditions in high dimension. To show the viability of our approach we discuss several case studies from insurance and finance.
    Date: 2021–09
  2. By: Dylan Brewer (School of Economics, Georgia Institute of Technology); Alyssa Carlson (Department of Economics, University of Missouri)
    Abstract: We study approaches for adjusting machine learning methods when the training sample differs from the prediction sample on unobserved dimensions. The machine learning literature predominately assumes selection only on observed dimensions. Common suggestions are to re-weight or control for variables that influence selection as solutions to selection on observables. Simulation results show that selection on unobservables increases mean squared prediction error using common machine-learning algorithms. Common machine learning practices such as re-weighting or controlling for variables that influence selection into the training or testing sample often worsens sample selection bias. We suggest two control-function approaches that remove the effects of selection bias before training and find that they reduce mean-squared prediction error in simulations with a high degree of selection. We apply these approaches to predicting the vote share of the incumbent in gubernatorial elections using previously observed re-election bids. We find that ignoring selection on unobservables leads to substantially higher predicted vote shares for the incumbent than when the control function approach is used.
    Keywords: sample selection, machine learning, control function, inverse probability weighting
    JEL: C13 C31 C55 D72
    Date: 2021–09
  3. By: Jascha Buchhorn; Berthold U. Wigger
    Abstract: Using neural networks, the present study replicates previous results on the prediction of student dropout obtained with decision trees and logistic regressions. For this purpose, multilayer perceptrons are trained on the same data as in the initial study. It is shown that neural networks lead to a significant improvement in the prediction of students at risk. Already after the first semester, potential dropouts can be identified with a probability of 95 percent.
    Keywords: neural networks, student dropout, replication study
    Date: 2021
  4. By: Lin Li
    Abstract: Financial trading aims to build profitable strategies to make wise investment decisions in the financial market. It has attracted interests in the machine learning community for a long time. This paper proposes to trade financial assets automatically using feature preprocessing skills and Recurrent Reinforcement Learning (RRL) algorithm. The strategy starts from technical indicators extracted from assets' market information. Then these technical indicators are preprocessed by Principal Component Analysis (PCA) and Discrete Wavelet Transform (DWT) and eventually inputted to the RRL algorithm to do the trading. The extensive empirical evidence shows that the proposed strategy is not only effective and robust in its performance, but also can mitigate the drawbacks underlying the initial trading using RRL.
    Date: 2021–09
  5. By: Isaac K. Ofori (University of Insubria, Varese, Italy)
    Abstract: A conspicuous lacuna in the literature on Sub-Saharan Africa (SSA) is the lack of clarity on variables key for driving and predicting inclusive growth. To address this, I train the machine learning algorithms for the Standard lasso, the Minimum Schwarz Bayesian Information Criterion (Minimum BIC) lasso, and the Adaptive lasso to study patterns in a dataset comprising 97 covariates of inclusive growth for 43 SSA countries. First, the regularization results show that only 13 variables are key for driving inclusive growth in SSA. Further, the results show that out of the 13, the poverty headcount (US$1.90) matters most. Second, the findings reveal that ‘Minimum BIC lasso’ is best for predicting inclusive growth in SSA. Policy recommendations are provided in line with the region’s green agenda and the coming into force of the African Continental Free Trade Area.
    Keywords: Clean Fuel, Economic Growth, Machine Learning, Lasso, Sub-Saharan Africa, Regularization, Poverty.
    JEL: C01 C14 C51 C52 C55 F43 O4 O55
    Date: 2021–01
  6. By: Alexander Jaax; Frédéric Gonzales; Annabelle Mourougane
    Abstract: The increasing importance of services trade in the global economy contrasts with the lack of timely data to monitor recent developments. The nowcasting models developed in this paper are aimed at providing insights into current changes in total services trade, as recorded in monthly statistics of the G7 countries. Combining machine-learning techniques and dynamic factor models, the methodology exploits traditional data and Google Trends search data. No single model outperforms the others, but a weighted average of the best models combining machine-learning with dynamic factor models seems to be a promising avenue. The best models improve one-step ahead predictive performance relative to a simple benchmark by 30-35% on average across G7 countries and trade flows. Nowcasting models are estimated to have captured about 67% of the fall in services exports due to the COVID-19 shock and 60% of the fall in imports on average across G7 economies.
    Keywords: Dynamic factor models, G7 economies, Machine learning
    JEL: C4 C22 F17
    Date: 2021–09–23
  7. By: Pihnastyi, Oleh; Sytnikova, Anastasiya
    Abstract: In this paper, the results of the model for forecasting the flow parameters of a distributed transport system of the conveyor type are briefly considered. It is shown that the model of the transport system based on the neural network can be successfully applied to predict the flow parameters of the transport system which consists of a very large number of sections.
    Keywords: conveyor; forecasting model; neural network
    JEL: C02 C14 C25 C44 D24 L23 Q21
    Date: 2021–09–03
  8. By: Jens Ludwig; Sendhil Mullainathan
    Abstract: Algorithms (in some form) are already widely used in the criminal justice system. We draw lessons from this experience for what is to come for the rest of society as machine learning diffuses. We find economists and other social scientists have a key role to play in shaping the impact of algorithms, in part through improving the tools used to build them.
    JEL: C01 C54 C55 D8 H0 K0
    Date: 2021–09
  9. By: Solveig Flaig; Gero Junike
    Abstract: In this research, we show how to expand existing approaches of generative adversarial networks (GANs) being used as economic scenario generators (ESG) to a whole internal model - with enough risk factors to model the full band-width of investments for an insurance company and for a one year horizon as required in Solvency 2. For validation of this approach as well as for optimisation of the GAN architecture, we develop new performance measures and provide a consistent, data-driven framework. Finally, we demonstrate that the results of a GAN-based ESG are similar to regulatory approved internal models in Europe. Therefore, GAN-based models can be seen as an assumption-free data-driven alternative way of market risk modelling.
    Date: 2021–09
  10. By: Nathan Ratledge; Gabriel Cadamuro; Brandon De la Cuesta; Matthieu Stigler; Marshall Burke
    Abstract: In many regions of the world, sparse data on key economic outcomes inhibits the development, targeting, and evaluation of public policy. We demonstrate how advancements in satellite imagery and machine learning can help ameliorate these data and inference challenges. In the context of an expansion of the electrical grid across Uganda, we show how a combination of satellite imagery and computer vision can be used to develop local-level livelihood measurements appropriate for inferring the causal impact of electricity access on livelihoods. We then show how ML-based inference techniques deliver more reliable estimates of the causal impact of electrification than traditional alternatives when applied to these data. We estimate that grid access improves village-level asset wealth in rural Uganda by 0.17 standard deviations, more than doubling the growth rate over our study period relative to untreated areas. Our results provide country-scale evidence on the impact of a key infrastructure investment, and provide a low-cost, generalizable approach to future policy evaluation in data sparse environments.
    JEL: O11 O18 Q01 Q4
    Date: 2021–09
  11. By: Pedro Salas-Rojo; Juan Gabriel Rodríguez
    Abstract: This paper explores how the inheritances received influence the distribution of wealth (financial, non-financial and total) in four developed ?but substantially different? countries: the United States, Canada, Italy and Spain. Following the inequality of opportunity literature, we first group individuals into types based on the inheritances received. Then, we estimate the between-types wealth inequality to approximate the part of overall wealth inequality explained by inheritances. After showing that traditional approaches lead to non-robust and arbitrary results, we apply Machine Learning methods to overcome this limitation. Among the available computing methods, we observe that the random forests is the most precise algorithm. By using this technique, we find that inheritances explain more than 65% of wealth inequality (Gini coefficient) in the US and Spain, and more than 40% in Italy and Canada. Finally, for the US and Italy, given the availability of parental education, we also include this circumstance in the analysis and study its interaction with inheritances. It is observed that the effect of inheritances is more prominent at the middle of the wealth distribution, while parental education is more important for the asset-poor.
    Keywords: Wealth inequality; inheritances; Machine Learning; inequality of opportunity; parental education
    JEL: C60 D31 D63 G51
    Date: 2020–12
  12. By: Leogrande, Angelo; Costantiello, Alberto
    Abstract: We estimate the relationships between innovation and human resources in Europe using the European Innovation Scoreboard of the European Commission for 36 countries for the period 2010-2019. We perform Panel Data with Fixed Effects, Random Effects, Pooled OLS, Dynamic Panel and WLS. We found that Human resources is positively associated to “Basic-school entrepreneurial education and training”, “Employment MHT manufacturing KIS services”, “Employment share Manufacturing (SD)”, “Lifelong learning”, “New doctorate graduates”, “R&D expenditure business sector”, “R&D expenditure public sector”, “Tertiary education”. Our results also show that “Human Resources” is negatively associated to “Government procurement of advanced technology products”, “Medium and high-tech product exports”, “SMEs innovating in-house”, “Venture capital”. In adjunct we perform a clusterization with k-Means algorithm and we find the presence of three clusters. Clusterization shows the presence of Central and Northern European countries that has higher levels of Human Resources, while Southern and Eastern Europe has very low degree of Human Resources. Finally, we use seven machine learning algorithms to predict the value of Human Resources in Europe Countries using data in the period 2014-2021 and we show that the linear regression algorithm performs at the highest level.
    Keywords: Innovation and Invention: Processes and Incentives, Management of Technological Innovation and R&D, Technological Change: Choices and Consequences, Diffusion Processes Intellectual Property and Intellectual Capital, Open Innovation, Government Policy.
    JEL: O30 O31 O32 O33 O34 O38
    Date: 2021–09
  13. By: Lin William Cong; Ke Tang; Bing Wang; Jingyuan Wang
    Abstract: We build a deep-learning-based SEIR-AIM model integrating the classical Susceptible-Exposed-Infectious-Removed epidemiology model with forecast modules of infection, community mobility, and unemployment. Through linking Google's multi-dimensional mobility index to economic activities, public health status, and mitigation policies, our AI-assisted model captures the populace's endogenous response to economic incentives and health risks. In addition to being an effective predictive tool, our analyses reveal that the long-term effective reproduction number of COVID-19 equilibrates around one before mass vaccination using data from the United States. We identify a "policy frontier" and identify reopening schools and workplaces to be the most effective. We also quantify protestors' employment-value-equivalence of the Black Lives Matter movement and find that its public health impact to be negligible.
    Date: 2021–09

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.