|
on Computational Economics |
|
Issue of 2019‒02‒04
sixteen papers chosen by |
| By: | Makarewicz, Tomasz |
| Abstract: | Behavioral and experimental literature on financial instability focuses on either subjective price expectations (Learning-to-Forecast experiments) or individual trading (Learning-to-Optimize experiments). Bao et al. (2017) have shown that subjects have problems with both tasks. In this paper, I explore these experimental results by investigating a model in which financial traders individually learn how to use forecasting and/or trading anchor-and-adjustment heuristics by updating them with Genetic Algorithms. The model replicates the main outcomes of these two threads of the experimental finance literature. It shows that both forecasters and traders coordinate on chasing asset price trends, which in turn causes substantial and self-fulfilling price oscillations, albeit larger and faster in the case of trading markets. When agents have to learn both tasks, financial instability becomes more persistent. |
| Keywords: | Financial Instability,Learning-to-Forecast and Learning-to-Optimize Experiments,Genetic Algorithm Model of Individual Learning |
| JEL: | C53 C63 C91 D03 D83 D84 |
| Date: | 2019 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:bamber:141&r=all |
| By: | Zhiyi Shen; Chengguo Weng |
| Abstract: | A number of optimal decision problems with uncertainty can be formulated into a stochastic optimal control framework. The Least-Squares Monte Carlo (LSMC) algorithm is a popular numerical method to approach solutions of such stochastic control problems as analytical solutions are not tractable in general. This paper generalizes the LSMC algorithm proposed in Shen and Weng (2017) to solve a wide class of stochastic optimal control models. Our algorithm has three pillars: a construction of auxiliary stochastic control model, an artificial simulation of the post-action value of state process, and a shape-preserving sieve estimation method which equip the algorithm with a number of merits including bypassing forward simulation and control randomization, evading extrapolating the value function, and alleviating computational burden of the tuning parameter selection. The efficacy of the algorithm is corroborated by an application to pricing equity-linked insurance products. |
| Date: | 2019–01 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1901.06715&r=all |
| By: | Blanka Horvath; Aitor Muguruza; Mehdi Tomas |
| Abstract: | We present a consistent neural network based calibration method for a number of volatility models -- including the rough volatility family -- that performs the calibration task within a few milliseconds for the full implied volatility surface. The aim of neural networks in this work is an off-line approximation of complex pricing functions, which are difficult to represent or time-consuming to evaluate by other means. We highlight how this perspective opens new horizons for quantitative modelling: The calibration bottleneck posed by a slow pricing of derivative contracts is lifted. This brings several model families (such as rough volatility models) within the scope of applicability in industry practice. As customary for machine learning, the form in which information from available data is extracted and stored is crucial for network performance. With this in mind, we discuss how our approach addresses the usual challenges of machine learning solutions in a financial context (availability of training data, interpretability of results for regulators, control over generalisation errors). We present specific architectures for price approximation and calibration and optimize these with respect to different objectives regarding accuracy, speed and robustness. We also find that including the intermediate step of learning pricing functions of (classical or rough) models before calibration significantly improves network performance compared to direct calibration to data. |
| Date: | 2019–01 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1901.09647&r=all |
| By: | Ermanno Catullo (Department of Economics and Social Sciences, Universita' Politecnica delle Marche); Federico Giri (Department of Economics and Social Sciences, Universita' Politecnica delle Marche); Mauro Gallegati (Department of Economics and Social Sciences, Universita' Politecnica delle Marche) |
| Abstract: | The paper presents an agent based model reproducing a stylized credit network that evolves endogenously through the individual choices of rms and banks. We introduce in this framework a anancial stability authority in order to test the e ects of different prudential policy measures designed to improve the resilience of the economic system. Simulations show that a combination of micro and macro prudential policies reduces systemic risk, but at the cost of increasing banks' capital volatility. Moreover, agent based methodology allows us to implement an alternative meso regulatory framework that takes into consideration the connections between firms and banks. This policy targets only the more connected banks, increasing their capital requirement in order to reduce the di usion of local shocks. Our results support the idea that the meso prudential policy is able to reduce systemic risk without a ecting the stability of banks'capital structure. |
| Keywords: | Micro prudential policy; Macro prudential policy; Credit Network; Meso prudential policy; Agent based model |
| JEL: | E50 E58 G18 G28 C63 |
| Date: | 2019–01 |
| URL: | http://d.repec.org/n?u=RePEc:anc:wpaper:434&r=all |
| By: | Giovanni Dosi (Laboratory of Economics and Management); Marcelo C. Pereira (Universidade Estadual de Campinas); Andrea Roventini (Observatoire français des conjonctures économiques); Maria Enrica Virgillito (Scuola Superiore Sant'Anna) |
| Abstract: | This work analyses the effects of labour market structural reforms by means of the labour-augmented ‘Schumpeter meeting Keynes’ (KþS) Agent-Based model. We introduce a policy regime change characterized by a set of structural reforms on the labour market. Confirming a recent IMF report, the model shows how structural reforms reducing workers’ bargaining power and compressing wages tend to increase (a) unemployment, (b) functional income inequality and (c) personal income inequality. We further undertake a global sensitivity analysis on key variables and parameters which corroborates the robustness of our findings. |
| Keywords: | Labor market; Structural reforms; Income distribution; Inequality; Unemployment; Growth |
| JEL: | C63 E02 E12 E24 O11 |
| Date: | 2017–12 |
| URL: | http://d.repec.org/n?u=RePEc:spo:wpmain:info:hdl:2441/3kbkotqp1b85pa2lu2puri38p6&r=all |
| By: | Kinne, Jan; Lenz, David |
| Abstract: | Innovation is considered as a main driver of economic growth. Promoting the development of innovation through STI (science, technology and innovation) policies requires accurate indicators of innovation. Traditional indicators often lack coverage, granularity as well as timeliness and involve high data collection costs, especially when conducted at a large scale. In this paper, we propose a novel approach on how to create firm-level innovation indicators at the scale of millions of firms. We use traditional firm-level innovation indicators from the questionnaire-based Community Innovation Survey (CIS) survey to train an artificial neural network classification model on labelled (innovative/non-innovative) web texts of surveyed firms. Subsequently, we apply this classification model to the web texts of hundreds of thousands of firms in Germany to predict their innovation status. Our results show that this approach produces credible predictions and has the potential to be a valuable and highly cost-efficient addition to the existing set of innovation indicators, especially due to its coverage and regional granularity. The predicted firm-level probabilities can also directly be interpreted as a continuous measure of innovativeness, opening up additional advantages over traditional binary innovation indicators. |
| Keywords: | Web Mining,Web Scraping,R&D,R&I,STI,Innovation,Indicators,Text Mining,Natural Language Processing,NLP,Deep Learning |
| JEL: | O30 C81 C83 |
| Date: | 2019 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:zewdip:19001&r=all |
| By: | Tommaso Ciarli (SPRU, University of Sussex, UK); Maria Savona (SPRU, University of Sussex, UK) |
| Abstract: | We discuss how different models assessing climate change integrate aspects of structural change that are crucial to improve understanding of the relation between changes in the environment and in the economy. We identify six related aspects of structural change, which have significant impact on climate change: sectoral composition, industrial organisation, technology, employment, final demand, and institutions. Economic models vary substantially with respect to the aspects of structural change that they include, and how they model them. We review different modelling families and compare these differences: integrated assessment models (IAM), computable general equilibrium (CGE) models, structural change models (SCM), ecological macroeconomics models in the Keynesian tradition (EMK) and evolutionary agent based models (EABM). We find that IAM and CGE address few of the aspects of structural change identified; SCM focus on the sectoral composition; and EKM study final demand and employment structure. But all these models are aggregate and omit the complexity of the interactions between structural and climate change. EABM have explored a larger number of aspects of structural change, modelling their emergence from the interaction of microeconomic actors, but have not yet exploited their potential to study the interactions among interrelated aspects of structural and climate change. |
| Keywords: | Strucural change, climate change, economic modelling |
| Date: | 2019–01 |
| URL: | http://d.repec.org/n?u=RePEc:sru:ssewps:2019-01&r=all |
| By: | M. Indra al Irsyad; Anthony Halog; Rabindra Nepal |
| Abstract: | This study estimates the impacts of four solar energy policy interventions on the photovoltaic (PV) market potential, government expenditure, economic growth, and the environment. An agent-based model is developed to capture the specific economic and institutional features of developing economies, citing Indonesia as a specific case study. We undertake a novel approach to energy modelling by combining energy system analysis, input-output analysis, life-cycle analysis, and socio-economic analysis to obtain a comprehensive and integrated impact assessment. Our results, after sensitivity analysis, call for abolishing the existing PV grant policy in the Indonesian rural electrification programs. The government, instead, should encourage the PV industry to improve production efficiency and to provide after-sales service. A 100-watt peak (Wp) PV under this policy is affordable for 33.2 percent of rural households without electricity access in 2010. Rural PV market size potentially increases to 82.4 percent with rural financing institutions lending 70 percent of capital cost for five years at 12 percent annual interest rate. Additional 30 percent capital subsidy and 5 percent interest subsidy slightly increase the rural PV market potential to 89.6 percent of PV adopters. However, the subsidies are crucial for creating PV demands by urban households but the most effective policy for promoting PV to urban households is the net metering scheme. Several policy proposals are discussed in response to these findings. |
| Keywords: | hybrid energy model, developing country, renewables policy, impact assessments, agent-based modelling, photovoltaic system |
| JEL: | C60 Q21 Q43 Q48 |
| Date: | 2019–01 |
| URL: | http://d.repec.org/n?u=RePEc:een:camaaa:2019-02&r=all |
| By: | Ensslen, Axel; Ringler, Philipp; Dörr, Lasse; Jochem, Patrick; Zimmermann, Florian; Fichtner, Wolf |
| Abstract: | Over the past few years, registration figures of plug-in electric vehicles have increased rapidly in industrialized countries. This could cause considerable mid- to long-term effects on electricity markets. To tackle potential challenges specific to electric power systems, we develop a load-shift-incentivizing electricity tariff that is suitable for electric vehicle users and analyze the tariff scheme in three parts. First, acceptance is analyzed based on surveys conducted among fleet managers and electric vehicle users. Corresponding results are used to calibrate the tariff. Secondly, load flexibilities of electric vehicle charging are used in an agent-based electricity market simulation model of the French and German wholesale electricity markets to simulate corresponding market impacts. Thirdly, the charging manager’s (‘aggregator’) business model is analyzed. Our results reveal that the tariff is highly suitable for incentivizing vehicle users to provide load flexibilities, which consequently increase the contribution margins of the charging managers. The main drawback is the potential for ‘avalanche effects’ on wholesale electricity markets increasing charging mangers’ expenditures, especially in France. |
| Keywords: | E-Mobility Electric vehicles Controlled charging Electricity markets |
| JEL: | O33 R42 |
| Date: | 2018–02–20 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:91543&r=all |
| By: | Nikolaos Passalis; Anastasios Tefas; Juho Kanniainen; Moncef Gabbouj; Alexandros Iosifidis |
| Abstract: | Time series forecasting is a crucial component of many important applications, ranging from forecasting the stock markets to energy load prediction. The high-dimensionality, velocity and variety of the data collected in these applications pose significant and unique challenges that must be carefully addressed for each of them. In this work, a novel Temporal Logistic Neural Bag-of-Features approach, that can be used to tackle these challenges, is proposed. The proposed method can be effectively combined with deep neural networks, leading to powerful deep learning models for time series analysis. However, combining existing BoF formulations with deep feature extractors pose significant challenges: the distribution of the input features is not stationary, tuning the hyper-parameters of the model can be especially difficult and the normalizations involved in the BoF model can cause significant instabilities during the training process. The proposed method is capable of overcoming these limitations by a employing a novel adaptive scaling mechanism and replacing the classical Gaussian-based density estimation involved in the regular BoF model with a logistic kernel. The effectiveness of the proposed approach is demonstrated using extensive experiments on a large-scale financial time series dataset that consists of more than 4 million limit orders. |
| Date: | 2019–01 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1901.08280&r=all |
| By: | Elena Green; Daniel M. Heffernan |
| Abstract: | This paper outlines an agent-based model of a simple financial market in which a single asset is available for trade by three different types of traders. The model was first introduced in the PhD thesis of one of the authors, see reference [1]. The simulated log returns are examined for the presence of the stylised facts of financial data. The features of leptokurtosis, volatility clustering and aggregational Gaussianity are especially highlighted and studied in detail. The following ingredients are found to be essential for the production of these stylised facts: the memory of noise traders who make random trade decisions; the inclusion of technical traders that trade in line with trends in the price and the inclusion of fundamental traders who know the "fundamental value" of the stock and trade accordingly. When these three basic types of traders are included log returns are produced with a leptokurtic distribution and volatility clustering as well as some further statistical features of empirical data. This enhances and broadens our understanding of the fundamental processes involved in the production of empirical data by the market. |
| Date: | 2019–01 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1901.05053&r=all |
| By: | M. Rouhani, Omid |
| Abstract: | In this paper, I briefly review the key methods to evaluate transportation projects. These methods are: Financial analysis; Cost benefit (economic analysis); Multi-criteria analysis; Cost-effectiveness analysis; Social welfare analysis; and Risk analysis (Monte Carlo simulation). The importance of understanding these methods lies in the fact that transportation projects offer huge social benefits and costs; some are impossible or very complex to measure in monetary terms. |
| Keywords: | Project evaluation methods, Transport projects, Social cost benefit analysis, Multi-criteria analysis, and Social welfare analysis. |
| JEL: | H43 R42 R58 |
| Date: | 2019–01–07 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:91451&r=all |
| By: | Steven Kivinen (Department of Economics, Dalhousie University) |
| Abstract: | I incorporate social networks into a search and matching model, allowing for congestion effects. The model predicts that the presence of network externalities increases the volatility of unemployment and other variables. I demonstrate analytically that aggregate matching functions exhibit decreasing returns to scale under certain conditions, that unemployment and matching rates have a larger response to productivity shocks, and that labour market tightness adjusts more slowly to its steady-state. Numerical simulations demonstrate that network effects can generate increases in the volatility of unemployment and matching rates, as well as increases in the autocorrelation of vacancies. |
| Keywords: | Social Networks; Unemployment; Search and Matching |
| Date: | 2017–03–15 |
| URL: | http://d.repec.org/n?u=RePEc:dal:wpaper:daleconwp2017-02&r=all |
| By: | Alda, Erik |
| Abstract: | This study examines the efficiency of local jails for the year 2016. It employs a well-known non-parametric methodology (DEA) with metafrontiers. Metafrontiers envelop separate groups that have similar production technology and therefore allows for more accurate efficiency estimates. Using an input-oriented model with variable returns to scale, the results of this study suggest that, on average, jails could reduce or reallocate their inputs by 37% given their output level. Also, there are differences in efficiency between groups that operate on different production technologies. The group of small jails appears to operate more efficiently than the groups of large and mega-large jails. From a managerial perspective, this study presents evidence that jail managers need to assess more carefully how they allocate human and financial resources to try to improve operational efficiency. From a policy perspective, the results indicate that there is room for cost-saving approaches to maximize taxpayer dollars. |
| Keywords: | Jails,Performance, Efficiency, Data Envelopment Analysis, Metafrontiers |
| JEL: | D61 |
| Date: | 2019–01–28 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:91803&r=all |
| By: | Hossein Sabzian; Mohammad Ali Shafia; Ali Maleki; Seyeed Mostapha Seyeed Hashemi; Ali Baghaei; Hossein Gharib |
| Abstract: | Nowadays, we are surrounded by a large number of complex phenomena ranging from rumor spreading, social norms formation to rise of new economic trends and disruption of traditional businesses. To deal with such phenomena,Complex Adaptive System (CAS) framework has been found very influential among social scientists,especially economists. As the most powerful methodology of CAS modeling, Agent-based modeling (ABM) has gained a growing application among academicians and practitioners. ABMs show how simple behavioral rules of agents and local interactions among them at micro-scale can generate surprisingly complex patterns at macro-scale. Despite a growing number of ABM publications, those researchers unfamiliar with this methodology have to study a number of works to understand (1) the why and what of ABMs and (2) the ways they are rigorously developed. Therefore, the major focus of this paper is to help social sciences researchers,especially economists get a big picture of ABMs and know how to develop them both systematically and rigorously. |
| Date: | 2019–01 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1901.08932&r=all |
| By: | Giovanni Dosi (Laboratory of Economics and Management); Marcelo C. Pereira (Universidade Estadual de Campinas); Andrea Roventini (Observatoire français des conjonctures économiques); Maria Enrica Virgillito (Scuola Superiore Sant'Anna) |
| Abstract: | In this work we discuss the research findings from the labour-augmented Schumpeter meeting Keynes (K+S) agent-based model. It comprises comparative dynamics experiments on an artificial economy populated by heterogeneous, interacting agents, as workers, firms, banks and the government. The exercises are characterised by different degrees of labour flexibility, or by institutional shocks entailing labour market structural reforms, wherein the phenomenon of hysteresis is endogenous and pervasive. The K+S model constitutes a laboratory to evaluate the effects of new institutional arrangements as active/passive labour market policies, and fiscal austerity. In this perspective, the model allows mimicking many of the customary policy responses which the European Union and many Latin American countries have embraced in reaction to the recent economic crises. The obtained results seem to indicate, however, that most of the proposed policies are likely inadequate to tackle the short-term crises consequences, and even risk demoting the long-run economic prospects. More objectively, the conclusions offer a possible explanation to the negative path traversed by economies like Brazil, where many of the mentioned policies were applied in a short period, and hint about some risks ahead. |
| Keywords: | Labour market ; Policy evaluation; Agent based model |
| JEL: | C63 E24 H53 J88 |
| Date: | 2018–09 |
| URL: | http://d.repec.org/n?u=RePEc:spo:wpmain:info:hdl:2441/5rtilga41c899ab0rctd3cp2r3&r=all |