nep-cmp New Economics Papers
on Computational Economics
Issue of 2020‒01‒13
thirty papers chosen by

  1. A Gated Recurrent Unit Approach to Bitcoin Price Prediction By Aniruddha Dutta; Saket Kumar; Meheli Basu
  2. Adversarial recovery of agent rewards from latent spaces of the limit order book By Jacobo Roa-Vicens; Yuanbo Wang; Virgile Mison; Yarin Gal; Ricardo Silva
  3. Solving dynamic discrete choice models using smoothing and sieve methods By Dennis Kristensen; Patrick K. Mogensen; Jong-Myun Moon; Bertel Schjerning
  4. Deep Learning for Decision Making and the Optimization of Socially Responsible Investments and Portfolio By Nhi N.Y.Vo; Xue-Zhong He; Shaowu Liu; Guandong Xu
  5. Computational optimization of associative learning experiments By Melinscak, Filip; Bach, Dominik R
  6. A new approach to Early Warning Systems for small European banks By Bräuning, Michael; Malikkidou, Despo; Scricco, Giorgio; Scalone, Stefano
  7. Corporate default forecasting with machine learning By Mirko Moscatelli; Simone Narizzano; Fabio Parlapiano; Gianluca Viggiano
  8. Sanction or Financial Crisis? An Artificial Neural Network-Based Approach to model the impact of oil price volatility on Stock and industry indices By Somayeh Kokabisaghi; Mohammadesmaeil Ezazi; Reza Tehrani; Nourmohammad Yaghoubi
  9. Deep Quarantine for Suspicious Mail By Benkovich, Nikita; Dedenok, Roman; Golubev, Dmitry
  10. Prior Knowledge Neural Network for Automatic Feature Construction in Financial Time Series By Jie Fang; Jianwu Lin; Yong Jiang; Shutao Xia
  11. Forecasting Bitcoin closing price series using linear regression and neural networks models By Nicola Uras; Lodovica Marchesi; Michele Marchesi; Roberto Tonelli
  12. Estimation of the yield curve for Costa Rica using combinatorial optimization metaheuristics applied to nonlinear regression By Andres Quiros-Granados; JAvier Trejos-Zelaya
  13. Fools Rush In: Competitive Effects of Reaction Time in Automated Trading By Henry Hanifan; John Cartlidge
  14. Multilevel Simulation of Demography and Food Production in Ancient Agrarian Societies: A Case Study from Roman North Africa By Gauthier, Nicolas
  15. Predicting intraday jumps in stock prices using liquidity measures and technical indicators By Ao Kong; Hongliang Zhu; Robert Azencott
  16. Efficient Algorithms for Constructing Multiplex Networks Embedding By Zolnikov, Pavel; Zubov, Maxim; Nikitinsky, Nikita; Makarov, Ilya
  17. Operator splitting schemes for American options under the two-asset Merton jump-diffusion model By Lynn Boen; Karel J. in 't Hout
  18. Deep Reinforcement Learning with VizDoomFirst-Person Shooter By Akimov, Dmitry; Makarov, Ilya
  19. Approximating intractable short ratemodel distribution with neural network By Anna Knezevic; Nikolai Dokuchaev
  20. Branch-and-Cut for the Active-Passive Vehicle Routing Problem By Christian Tilk; Michael Forbes
  21. Fast Value Iteration: An Application of Legendre-Fenchel Duality to a Class of Deterministic Dynamic Programming Problems in Discrete Time By Ronaldo Carpio; Takashi Kamihigashi
  22. Online Quantification of Input Model Uncertainty by Two-Layer Importance Sampling By Tianyi Liu; Enlu Zhou
  23. Adaptive Discrete Smoothing for High-Dimensional and Nonlinear Panel Data By Xi Chen; Victor Chernozhukov; Ye Luo; Martin Spindler
  24. A survey on kriging-based infill algorithms for multiobjective simulation optimization By Sebastian Rojas Gonzalez; Inneke Van Nieuwenhuyse
  25. Energy Scenario Exploration with Modeling to Generate Alternatives (MGA) By Joseph F. DeCarolis; Samaneh Babaee; Binghui Li; Suyash Kanungo
  26. General Game Playing B-to-B Price Negotiations By Michael, Friedrich; Ignatov, Dmitry I.
  27. FAIRNESS MEETS MACHINE LEARNING: SEARCHING FOR A BETTER BALANCE By Ekaterina Semenova; Ekaterina Perevoshchikova; Alexey Ivanov; Mikhail Erofeev
  28. Fairness in Multi-agent Reinforcement Learning for Stock Trading By Wenhang Bao
  29. Finite Horizon Dynamic Games with and without a Scrap Value By Reinhard Neck; Dmitri Blueschke; Viktoria Blueschke-Nikolaeva
  30. How do machine learning and non-traditional data affect credit scoring? New evidence from a Chinese fintech firm By Leonardo Gambacorta; Yiping Huang; Han Qiu; Jingyi Wang

  1. By: Aniruddha Dutta; Saket Kumar; Meheli Basu
    Abstract: In today's era of big data, deep learning and artificial intelligence have formed the backbone for cryptocurrency portfolio optimization. Researchers have investigated various state of the art machine learning models to predict Bitcoin price and volatility. Machine learning models like recurrent neural network (RNN) and long short-term memory (LSTM) have been shown to perform better than traditional time series models in cryptocurrency price prediction. However, very few studies have applied sequence models with robust feature engineering to predict future pricing. in this study, we investigate a framework with a set of advanced machine learning methods with a fixed set of exogenous and endogenous factors to predict daily Bitcoin prices. We study and compare different approaches using the root mean squared error (RMSE). Experimental results show that gated recurring unit (GRU) model with recurrent dropout performs better better than popular existing models. We also show that simple trading strategies, when implemented with our proposed GRU model and with proper learning, can lead to financial gain.
    Date: 2019–12
  2. By: Jacobo Roa-Vicens; Yuanbo Wang; Virgile Mison; Yarin Gal; Ricardo Silva
    Abstract: Inverse reinforcement learning has proved its ability to explain state-action trajectories of expert agents by recovering their underlying reward functions in increasingly challenging environments. Recent advances in adversarial learning have allowed extending inverse RL to applications with non-stationary environment dynamics unknown to the agents, arbitrary structures of reward functions and improved handling of the ambiguities inherent to the ill-posed nature of inverse RL. This is particularly relevant in real time applications on stochastic environments involving risk, like volatile financial markets. Moreover, recent work on simulation of complex environments enable learning algorithms to engage with real market data through simulations of its latent space representations, avoiding a costly exploration of the original environment. In this paper, we explore whether adversarial inverse RL algorithms can be adapted and trained within such latent space simulations from real market data, while maintaining their ability to recover agent rewards robust to variations in the underlying dynamics, and transfer them to new regimes of the original environment.
    Date: 2019–12
  3. By: Dennis Kristensen (Institute for Fiscal Studies and University College London); Patrick K. Mogensen (Institute for Fiscal Studies); Jong-Myun Moon (Institute for Fiscal Studies and University College London); Bertel Schjerning (Institute for Fiscal Studies and University of Copenhagen)
    Abstract: We propose to combine smoothing, simulations and sieve approximations to solve for either the integrated or expected value function in a general class of dynamic discrete choice (DDC) models. We use importance sampling to approximate the Bellman operators defining the two functions. The random Bellman operators, and therefore also the corresponding solutions, are generally non-smooth which is undesirable. To circumvent this issue, we introduce a smoothed version of the random Bellman operator and solve for the corresponding smoothed value function using sieve methods. We show that one can avoid using sieves by generalizing and adapting the “self-approximating” method of Rust (1997b) to our setting. We provide an asymptotic theory for the approximate solutions and show that they converge with vN-rate, where N is number of Monte Carlo draws, towards Gaussian processes. We examine their performance in practice through a set of numerical experiments and find that both methods perform well with the sieve method being particularly attractive in terms of computational speed and accuracy.
    Keywords: Dynamic discrete choice; numerical solution; Monte Carlo; sieves
    Date: 2019–04–03
  4. By: Nhi N.Y.Vo; Xue-Zhong He (Finance Discipline Group, University of Technology Sydney); Shaowu Liu; Guandong Xu
    Abstract: A socially responsible investment portfolio takes into consideration the environmental, social and governance aspects of companies. It has become an emerging topic for both financial investors and researchers recently. Traditional investment and portfolio theories, which are used for the optimization of financial investment portfolios, are inadequate for decision-making and the construction of an optimized socially responsible investment portfolio. In response to this problem, we introduced a Deep Responsible Investment Portfolio (DRIP) model that contains a Multivariate Bidirectional Long Short-Term Memory neural network, to predict stock returns for the construction of a socially responsible investment portfolio. The deep reinforcement learning technique was adapted to retrain neural networks and rebalance the portfolio periodically. Our empirical data revealed that the DRIP framework could achieve competitive financial performance and better social impact compared to traditional portfolio models, sustainable indexes and funds.
    Keywords: Socially responsible investment; Portfolio optimization; Multivariate analytics; Deep reinforcement learning; Decision support systems
    Date: 2019–01–01
  5. By: Melinscak, Filip; Bach, Dominik R
    Abstract: With computational biology striving to provide more accurate theoretical accounts of biological systems, use of increasingly complex computational models seems inevitable. However, this trend engenders a challenge of optimal experimental design: due to the flexibility of complex models, it is difficult to intuitively design experiments that will efficiently expose differences between candidate models or allow accurate estimation of their parameters. This challenge is well exemplified in associative learning research. Associative learning theory has a rich tradition of computational modeling, resulting in a growing space of increasingly complex models, which in turn renders manual design of informative experiments difficult. Here we propose a novel method for computational optimization of associative learning experiments. We first formalize associative learning experiments using a low number of tunable design variables, to make optimization tractable. Next, we combine simulation-based Bayesian experimental design with Bayesian optimization to arrive at a flexible method of tuning design variables. Finally, we validate the proposed method through extensive simulations covering both the objectives of accurate parameter estimation and model selection. The validation results show that computationally optimized experimental designs have the potential to substantially improve upon manual designs drawn from the literature, even when prior information guiding the optimization is scarce. Computational optimization of experiments may help address recent concerns over reproducibility by increasing the expected utility of studies, and it may even incentivize practices such as study pre-registration, since optimization requires a pre-specified analysis plan. Moreover, design optimization has the potential not only to improve basic research in domains such as associative learning, but also to play an important role in translational research. For example, design of behavioral and physiological diagnostic tests in the nascent field of computational psychiatry could benefit from an optimization-based approach, similar to the one presented here.
    Date: 2019–02–20
  6. By: Bräuning, Michael; Malikkidou, Despo; Scricco, Giorgio; Scalone, Stefano
    Abstract: This paper describes a machine learning technique to timely identify cases of individual bank financial distress. Our work represents the first attempt in the literature to develop an early warning system specifically for small European banks. We employ a machine learning technique, and build a decision tree model using a dataset of official supervisory reporting, complemented with qualitative banking sector and macroeconomic variables. We propose a new and wider definition of financial distress, in order to capture bank distress cases at an earlier stage with respect to the existing literature on bank failures; by doing so, given the rarity of bank defaults in Europe we significantly increase the number of events on which to estimate the model, thus increasing the model precision; in this way we identify bank crises at an earlier stage with respect to the usual default definition, therefore leaving a time window for supervisory intervention. The Quinlan C5.0 algorithm we use to estimate the model also allows us to adopt a conservative approach to misclassification: as we deal with bank distress cases, we consider missing a distress event twice as costly as raising a false flag. Our final model comprises 12 variables in 19 nodes, and outperforms a logit model estimation, which we use to benchmark our analysis; validation and back testing also suggest that the good performance of our model is relatively stable and robust. JEL Classification: E58, C01, C50
    Keywords: bank distress, decision tree, machine learning, Quinlan
    Date: 2019–12
  7. By: Mirko Moscatelli (Bank of Italy); Simone Narizzano (Bank of Italy); Fabio Parlapiano (Bank of Italy); Gianluca Viggiano (Bank of Italy)
    Abstract: We analyze the performance of a set of machine learning (ML) models in predicting default risk, using standard statistical models, such as the logistic regression, as a benchmark. When only a limited information set is available, for example in the case of financial indicators, we find that ML models provide substantial gains in discriminatory power and precision compared with statistical models. This advantage diminishes when high quality information, such as credit behavioral indicators obtained from the Credit Register, is also available, and becomes negligible when the dataset is small. We also evaluate the consequences of using an ML-based rating system on the supply of credit and the number of borrowers gaining access to credit. ML models channel a larger share of credit towards safer and larger borrowers and result in lower credit losses for lenders.
    Keywords: Credit Scoring, Machine Learning, Random Forest, Gradient Boosting Machine
    JEL: G2 C52 C55 D83
    Date: 2019–12
  8. By: Somayeh Kokabisaghi; Mohammadesmaeil Ezazi; Reza Tehrani; Nourmohammad Yaghoubi
    Abstract: Financial market in oil-dependent countries has been always influenced by any changes in international energy market, In particular, oil price.It is therefore of considerable interest to investigate the impact of oil price on financial markets. The aim of this paper is to model the impact of oil price volatility on stock and industry indices by considering gas and gold price,exchange rate and trading volume as explanatory variables. We also propose Feed-forward networks as an accurate method to model non-linearity. we use data from 2009 to 2018 that is split in two periods during international energy sanction and post-sanction. The results show that Feed-forward networks perform well in predicting variables and oil price volatility has a significant impact on stock and industry market indices. The result is more robust in the post-sanction period and global financial crisis in 2014. Herein, it is important for financial market analysts and policy makers to note which factors and when influence the financial market, especially in an oil-dependent country such as Iran with uncertainty in the international politics. This research analyses the results in two different periods, which is important in the terms of oil price shock and international energy sanction. Also, using neural networks in methodology gives more accurate and reliable results. Keywords: Feed-forward networks,Industry index,International energy sanction,Oil price volatility
    Date: 2019–12
  9. By: Benkovich, Nikita; Dedenok, Roman; Golubev, Dmitry
    Abstract: In this paper, we introduce DeepQuarantine (DQ), a cloudtechnology to detect and quarantine potential spam messages. Spam at-tacks are becoming more diverse and can potentially be harmful to emailusers. Despite the high quality and performance of spam filtering sys-tems, detection of a spam campaign can take some time. Unfortunately,in this case some unwanted messages get delivered to users. To solve thisproblem, we created DQ, which detects potential spam and keeps it ina special Quarantine folder for a while. The time gained allows us todouble-check the messages to improve the reliability of the anti-spam so-lution. Due to high precision of the technology, most of the quarantinedmail is spam, which allows clients to use email without delay. Our solutionis based on applying Convolutional Neural Networks on MIME headersto extract deep features from large-scale historical data. We evaluatedthe proposed method on real-world data and showed that DQ enhancesthe quality of spam detection.
    Keywords: spam filtering; spam detection; machine learning; deeplearning; cloud technology
    JEL: C45 M15
    Date: 2019–09–23
  10. By: Jie Fang; Jianwu Lin; Yong Jiang; Shutao Xia
    Abstract: In quantitative finance, useful features are constructed by human experts. However, this method is of low efficient. Thus, automatic feature construction algorithms have received more and more attention. The SOTA technic in this field is to use reverse polish expression to represent the features, and then use genetic programming to reconstruct it. In this paper, we propose a new method, alpha discovery neural network, which can automatically construct features by using neural network. In this work, we made several contributions. Firstly, we put forward new object function by using empirical knowledge in financial signal processing, and we also fixed its undifferentiated problem. Secondly, we use model stealing technic to learn from other prior knowledge, which can bring enough diversity into our network. Thirdly, we come up with a method to measure the diversity of different financial features. Experiment shows that ADN can produce more diversified and higher informative features than GP. Besides, if we use GP's output to serve as prior knowledge, its final achievements will be significantly improved by using ADN.
    Date: 2019–12
  11. By: Nicola Uras; Lodovica Marchesi; Michele Marchesi; Roberto Tonelli
    Abstract: This paper studies how to forecast daily closing price series of Bitcoin, using data on prices and volumes of prior days. Bitcoin price behaviour is still largely unexplored, presenting new opportunities. We compared our results with two modern works on Bitcoin prices forecasting and with a well-known recent paper that uses Intel, National Bank shares and Microsoft daily NASDAQ closing prices spanning a 3-year interval. We followed different approaches in parallel, implementing both statistical techniques and machine learning algorithms. The SLR model for univariate series forecast uses only closing prices, whereas the MLR model for multivariate series uses both price and volume data. We applied the ADF -Test to these series, which resulted to be indistinguishable from a random walk. We also used two artificial neural networks: MLP and LSTM. We then partitioned the dataset into shorter sequences, representing different price regimes, obtaining best result using more than one previous price, thus confirming our regime hypothesis. All the models were evaluated in terms of MAPE and relativeRMSE. They performed well, and were overall better than those obtained in the benchmarks. Based on the results, it was possible to demonstrate the efficacy of the proposed methodology and its contribution to the state-of-the-art.
    Date: 2020–01
  12. By: Andres Quiros-Granados; JAvier Trejos-Zelaya
    Abstract: The term structure of interest rates or yield curve is a function relating the interest rate with its own term. Nonlinear regression models of Nelson-Siegel and Svensson were used to estimate the yield curve using a sample of historical data supplied by the National Stock Exchange of Costa Rica. The optimization problem involved in the estimation process of model parameters is addressed by the use of four well known combinatorial optimization metaheuristics: Ant colony optimization, Genetic algorithm, Particle swarm optimization and Simulated annealing. The aim of the study is to improve the local minima obtained by a classical quasi-Newton optimization method using a descent direction. Good results with at least two metaheuristics are achieved, Particle swarm optimization and Simulated annealing. Keywords: Yield curve, nonlinear regression, Nelson-
    Date: 2019–11
  13. By: Henry Hanifan; John Cartlidge
    Abstract: We explore the competitive effects of reaction time of automated trading strategies in simulated financial markets containing a single exchange with public limit order book and continuous double auction matching. A large body of research conducted over several decades has been devoted to trading agent design and simulation, but the majority of this work focuses on pricing strategy and does not consider the time taken for these strategies to compute. In real-world financial markets, speed is known to heavily influence the design of automated trading algorithms, with the generally accepted wisdom that faster is better. Here, we introduce increasingly realistic models of trading speed and profile the computation times of a suite of eminent trading algorithms from the literature. Results demonstrate that: (a) trading performance is impacted by speed, but faster is not always better; (b) the Adaptive-Aggressive (AA) algorithm, until recently considered the most dominant trading strategy in the literature, is outperformed by the simplistic Shaver (SHVR) strategy - shave one tick off the current best bid or ask - when relative computation times are accurately simulated.
    Date: 2019–12
  14. By: Gauthier, Nicolas (University of Arizona)
    Abstract: Feedbacks between population growth, food production, and the environment were central to the growth and decay of ancient agrarian societies. Population growth increases both the number of mouths a society must feed and the number of people working to feed them. The balance between these two forces depends on the population's age structure. Although age structure ultimately reflects individual fertility and mortality, it is households that make decisions about the production and consumption of food, and their decisions depend on interactions with all other households in a settlement. How do these organizational levels interact to influence population growth and regulation? Here, I present a multi-level agent-based model of demography, food production, and social interaction in agricultural societies. I use the model to simulate the interactions of individuals, households, and settlements in a food-limited environment, and investigate the resulting patterns of population growth. Using Roman North Africa as a motivating example, I illustrate how abstract properties like "carrying capacity" emerge from the concrete actions and interactions of millions of individual people. Looking forward, bottom-up simulations rooted in first principles of human behavior will be crucial for understanding the coevolution of preindustrial societies and their natural environments.
    Date: 2019–08–22
  15. By: Ao Kong; Hongliang Zhu; Robert Azencott
    Abstract: Predicting the intraday stock jumps is a significant but challenging problem in finance. Due to the instantaneity and imperceptibility characteristics of intraday stock jumps, relevant studies on their predictability remain limited. This paper proposes a data-driven approach to predict intraday stock jumps using the information embedded in liquidity measures and technical indicators. Specifically, a trading day is divided into a series of 5-minute intervals, and at the end of each interval, the candidate attributes defined by liquidity measures and technical indicators are input into machine learning algorithms to predict the arrival of a stock jump as well as its direction in the following 5-minute interval. Empirical study is conducted on the level-2 high-frequency data of 1271 stocks in the Shenzhen Stock Exchange of China to validate our approach. The result provides initial evidence of the predictability of jump arrivals and jump directions using level-2 stock data as well as the effectiveness of using a combination of liquidity measures and technical indicators in this prediction. We also reveal the superiority of using random forest compared to other machine learning algorithms in building prediction models. Importantly, our study provides a portable data-driven approach that exploits liquidity and technical information from level-2 stock data to predict intraday price jumps of individual stocks.
    Date: 2019–12
  16. By: Zolnikov, Pavel; Zubov, Maxim; Nikitinsky, Nikita; Makarov, Ilya
    Abstract: Network embedding has become a very promising techniquein analysis of complex networks. It is a method to project nodes of anetwork into a low-dimensional vector space while retaining the structureof the network based on vector similarity. There are many methods ofnetwork embedding developed for traditional single layer networks. Onthe other hand, multilayer networks can provide more information aboutrelationships between nodes. In this paper, we present our random walkbased multilayer network embedding and compare it with single layerand multilayer network embeddings. For this purpose, we used severalclassic datasets usually used in network embedding experiments and alsocollected our own dataset of papers and authors indexed in Scopus.
    Keywords: Network embedding; Multi-layer network; Machine learning on graphs
    JEL: C45 I20
    Date: 2019–09–23
  17. By: Lynn Boen; Karel J. in 't Hout
    Abstract: This paper deals with the efficient numerical solution of the two-dimensional partial integro-differential complementarity problem (PIDCP) that holds for the value of American-style options under the two-asset Merton jump-diffusion model. We consider the adaptation of various operator splitting schemes of both the implicit-explicit (IMEX) and the alternating direction implicit (ADI) kind that have recently been studied for partial integro-differential equations (PIDEs) in [3]. Each of these schemes conveniently treats the nonlocal integral part in an explicit manner. Their adaptation to PIDCPs is achieved through a combination with the Ikonen-Toivanen splitting technique [14] as well as with the penalty method [32]. The convergence behaviour and relative performance of the acquired eight operator splitting methods is investigated in extensive numerical experiments for American put-on-the-min and put-on-the-average options.
    Date: 2019–12
  18. By: Akimov, Dmitry; Makarov, Ilya
    Abstract: In this work, we study deep reinforcement algorithms forpartially observable Markov decision processes (POMDP) combined withDeep Q-Networks. To our knowledge, we are the first to apply standardMarkov decision process architectures to POMDP scenarios. We proposean extension of DQN with Dueling Networks and several other model-freepolicies to training agent using deep reinforcement learning in VizDoomenvironment, which is replication of Doom first-person shooter. We de-velop several agents for the following scenarios in VizDoom first-personshooter (FPS): Basic, Defend The Center, Health Gathering. We com-pare our agent with Recurrent DQN with Prioritized Experience Replayand Snapshot Ensembling agent and get approximately triple increase inper episode reward. It is important to say that POMDP scenario closethe gap between human and computer player scenarios thus providingmore meaningful justification for Deep RL agent performance.
    Keywords: Deep Reinforcement Learning; VizDoom; First-Person Shooter; DQN; Double Q-learning; Dueling
    JEL: C02 C63 C88
    Date: 2019–09–23
  19. By: Anna Knezevic; Nikolai Dokuchaev
    Abstract: We propose an algorithm which predicts each subsequent time step relative to the previous time step of intractable short rate model (when adjusted for drift and overall distribution of previous percentile result) and show that the method achieves superior outcomes to the unbiased estimate both on the trained dataset and different validation data.
    Date: 2019–12
  20. By: Christian Tilk (Johannes Gutenberg University Mainz); Michael Forbes (The University of Queensland St Lucia)
    Abstract: This paper studies the active-passive vehicle-routing problem (APVRP).The APVRP coversarange of logistics applications where pickup-and-delivery requests necessitate a joint operation of active vehicles (e.g., trucks) and passive vehicles (e.g., loading devices such as containers). It supports a flexible coupling and decoupling of active and passive vehicles at customer locations in order to achieve a high utilization of both resources. This flexibility raises the need to synchronize the operations and the movements of active and passive vehicles in time and space. The contribution of the paper is twofold. First, we present a branch-and-cut algorithm for the exact solution of the APVRP that is based on Benders decomposition. Second, our approach can be generalized to deal with other vehicle-routing problems with timing aspects and synchronization constraints. Especially for the more complicated cases in which completion time or duration of routes is part of the objective, we show how stronger optimality cuts can be deï¬ ned by identifying minimal responsible subset. Computational experiments show that the proposed algorithm outperforms the previousstate-of-the-artfortheAPVRPandcomputeoptimalsolutionsformorethan70previously unsolved benchmark instances.
    Keywords: Routing, synchronization, branch-and-cut, Benders decomposition, truck and trailer
    Date: 2019–12–09
  21. By: Ronaldo Carpio (School of Business and Finance, University of International Business and Economics); Takashi Kamihigashi (Research Institute for Economics and Business Administration, Kobe University)
    Abstract: We propose an algorithm, which we call "Fast Value Iteration" (FVI), to compute the value function of a deterministic infinite-horizon dynamic programming problem in discrete time. FVI is an ecient algorithm applicable to a class of multidimen- sional dynamic programming problems with concave return (or convex cost) functions and linear constraints. In this algorithm, a sequence of functions is generated starting from the zero function by repeatedly applying a simple algebraic rule involving the Legendre-Fenchel transform of the return function. The resulting sequence is guaran- teed to converge, and the Legendre-Fenchel transform of the limiting function coincides with the value function.
    Keywords: Dynamic programming, Legendre-Fenchel transform, Bellman operator, Convex analysis
    Date: 2019–12
  22. By: Tianyi Liu; Enlu Zhou
    Abstract: Stochastic simulation has been widely used to analyze the performance of complex stochastic systems and facilitate decision making in those systems. Stochastic simulation is driven by the input model, which is a collection of probability distributions that model the stochasticity in the system. The input model is usually estimated using a finite amount of data, which introduces the so-called input model uncertainty (or, input uncertainty for short) to the simulation output. How to quantify input uncertainty has been studied extensively, and many methods have been proposed for the batch data setting, i.e., when all the data are available at once. However, methods for ``streaming data'' arriving sequentially in time are still in demand, despite that streaming data have become increasingly prevalent in modern applications. To fill in this gap, we propose a two-layer importance sampling framework that incorporates streaming data for online input uncertainty quantification. Under this framework, we develop two algorithms that suit two different application scenarios: the first is when data come at a fast speed and there is no time for any simulation in between updates; the second is when data come at a moderate speed and a few but limited simulations are allowed at each time stage. We show the consistency and asymptotic convergence rate results, which theoretically show the efficiency of our proposed approach. We further demonstrate the proposed algorithms on an example of the news vendor problem.
    Date: 2019–12
  23. By: Xi Chen; Victor Chernozhukov; Ye Luo; Martin Spindler
    Abstract: In this paper we develop a data-driven smoothing technique for high-dimensional and non-linear panel data models. We allow for individual specific (non-linear) functions and estimation with econometric or machine learning methods by using weighted observations from other individuals. The weights are determined by a data-driven way and depend on the similarity between the corresponding functions and are measured based on initial estimates. The key feature of such a procedure is that it clusters individuals based on the distance / similarity between them, estimated in a first stage. Our estimation method can be combined with various statistical estimation procedures, in particular modern machine learning methods which are in particular fruitful in the high-dimensional case and with complex, heterogeneous data. The approach can be interpreted as a \textquotedblleft soft-clustering\textquotedblright\ in comparison to traditional\textquotedblleft\ hard clustering\textquotedblright that assigns each individual to exactly one group. We conduct a simulation study which shows that the prediction can be greatly improved by using our estimator. Finally, we analyze a big data set from, a leading company in transportation industry, to analyze and predict the gap between supply and demand based on a large set of covariates. Our estimator clearly performs much better in out-of-sample prediction compared to existing linear panel data estimators.
    Date: 2019–12
  24. By: Sebastian Rojas Gonzalez; Inneke Van Nieuwenhuyse
    Date: 2019–03
  25. By: Joseph F. DeCarolis; Samaneh Babaee; Binghui Li; Suyash Kanungo
    Abstract: Energy system optimization models (ESOMs) should be used in an interactive way to uncover knife-edge solutions, explore alternative system configurations, and suggest different ways to achieve policy objectives under conditions of deep uncertainty. In this paper, we do so by employing an existing optimization technique called modeling to generate alternatives (MGA), which involves a change in the model structure in order to systematically explore the near-optimal decision space. The MGA capability is incorporated into Tools for Energy Model Optimization and Analysis (Temoa), an open source framework that also includes a technology rich, bottom up ESOM. In this analysis, Temoa is used to explore alternative energy futures in a simplified single region energy system that represents the U.S. electric sector and a portion of the light duty transport sector. Given the dataset limitations, we place greater emphasis on the methodological approach rather than specific results.
    Date: 2019–12
  26. By: Michael, Friedrich; Ignatov, Dmitry I.
    Abstract: This papers discusses the scientific and practical perspectives of using general game playing in business-to-business price negotiations as a part of Procurement 4.0 revolution. The status quo of digital price negotiations software,which emerged from intuitive solutions to business goals and refereed to as electronic auctions in industry, is summarized in scientific context. Description of such aspects as auctioneers’ interventions, asymmetry among players and time-depended features reveals the nature of nowadays electronic auctions to be rather termed as price games. This paper strongly suggests general game playing as the crucial technology for automation of human rule setting in those games. Game theory, genetic programming, experimental economics and AI human player simulation are also discussed as satellite topics. SIDL-type game descriptions languages and their formal game theoretic foundations are presented.
    Keywords: Procurement 4.0; Artificial Intelligence; General Game Playing; Game Theory; Mechanism Design; Experimental Economics; Behavioral Eco-nomics; z-Tree; Cognitive Modeling; e-Auctions; barter double auction; B-to-B Price Negotiations; English Auction; Dutch auction; Sealed-Bid Auction; Industry 4.0
    JEL: C63 C72 C90 D04 D44
    Date: 2019–09–23
  27. By: Ekaterina Semenova (National Research University Higher School of Economics); Ekaterina Perevoshchikova (National Research University Higher School of Economics); Alexey Ivanov (National Research University Higher School of Economics); Mikhail Erofeev (Lomonosov Moscow State University)
    Abstract: Machine learning (ML) affects nearly every aspect of our lives, including the weightiest ones such as criminal justice. As it becomes more widespread, however, it raises the question of how we can integrate fairness into ML algorithms to ensure that all citizens receive equal treatment and to avoid imperiling society’s democratic values. In this paper we study various formal definitions of fairness that can be embedded into ML algorithms and show that the root cause of most debates about AI fairness is society’s lack of a consistent understanding of fairness generally. We conclude that AI regulations stipulating an abstract fairness principle are ineffective societally. Capitalizing on extensive related work in computer science and the humanities, we present an approach that can help ML developers choose a formal definition of fairness suitable for a particular country and application domain. Abstract rules from the human world fail in the ML world and ML developers will never be free from criticism if the status quo remains. We argue that the law should shift from an abstract definition of fairness to a formal legal definition. Legislators and society as a whole should tackle the challenge of defining fairness, but since no definition perfectly matches the human sense of fairness, legislators must publicly acknowledge the drawbacks of the chosen definition and assert that the benefits outweigh them. Doing so creates transparent standards of fairness to ensure that technology serves the values and best interests of society
    Keywords: Artificial Intelligence; Bias; Fairness; Machine Learning; Regulation; Values; Antidiscrimination Law;
    JEL: K19
    Date: 2019
  28. By: Wenhang Bao
    Abstract: Unfair stock trading strategies have been shown to be one of the most negative perceptions that customers can have concerning trading and may result in long-term losses for a company. Investment banks usually place trading orders for multiple clients with the same target assets but different order sizes and diverse requirements such as time frame and risk aversion level, thereby total earning and individual earning cannot be optimized at the same time. Orders executed earlier would affect the market price level, so late execution usually means additional implementation cost. In this paper, we propose a novel scheme that utilizes multi-agent reinforcement learning systems to derive stock trading strategies for all clients which keep a balance between revenue and fairness. First, we demonstrate that Reinforcement learning (RL) is able to learn from experience and adapt the trading strategies to the complex market environment. Secondly, we show that the Multi-agent RL system allows developing trading strategies for all clients individually, thus optimizing individual revenue. Thirdly, we use the Generalized Gini Index (GGI) aggregation function to control the fairness level of the revenue across all clients. Lastly, we empirically demonstrate the superiority of the novel scheme in improving fairness meanwhile maintaining optimization of revenue.
    Date: 2019–12
  29. By: Reinhard Neck (Alpen-Adria-Universität Klagenfurt); Dmitri Blueschke (Alpen-Adria-Universität Klagenfurt); Viktoria Blueschke-Nikolaeva (Alpen-Adria-Universität Klagenfurt)
    Abstract: In this paper, we examine the effects of scrap values on the solutions of dynamic game Problems with a finite time horizon. We show how to include a scrap value in the OPTGAME3 algorithm for the numerical calculation of solutions for dynamic games. We consider two alternative ways of including a scrap value, either only for the state variables or for both the state and control variables. Using a numerical macroeconomic model of a monetary union, we show that the introduction of a scrap value is not appropriate as a substitute for an infinite horizon in dynamic economic policy game problems.
    Keywords: dynamic games, scrap value, finite horizon, Pareto solution, feedback Nash equilibrium solution
    JEL: C73 E60
    Date: 2019–10
  30. By: Leonardo Gambacorta; Yiping Huang; Han Qiu; Jingyi Wang
    Abstract: This paper compares the predictive power of credit scoring models based on machine learning techniques with that of traditional loss and default models. Using proprietary transaction-level data from a leading fintech company in China for the period between May and September 2017, we test the performance of different models to predict losses and defaults both in normal times and when the economy is subject to a shock. In particular, we analyse the case of an (exogenous) change in regulation policy on shadow banking in China that caused lending to decline and credit conditions to deteriorate. We find that the model based on machine learning and non-traditional data is better able to predict losses and defaults than traditional models in the presence of a negative shock to the aggregate credit supply. One possible reason for this is that machine learning can better mine the non-linear relationship between variables in a period of stress. Finally, the comparative advantage of the model that uses the fintech credit scoring technique based on machine learning and big data tends to decline for borrowers with a longer credit history.
    Keywords: fintech, credit scoring, non-traditional information, machine learning, credit risk
    JEL: G17 G18 G23 G32
    Date: 2019–12

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.