|
on Computational Economics |
|
Issue of 2018‒04‒23
fifteen papers chosen by |
| By: | Rongju Zhang; Nicolas Langren\'e; Yu Tian; Zili Zhu; Fima Klebaner; Kais Hamza |
| Abstract: | The least squares Monte Carlo algorithm has become a popular tool for solving stochastic control problems, mainly due to its ability to handle multiple stochastic state variables. However, a configuration that remains challenging is when dealing with high dimensional controls, due to the exponential complexity of grid search methods or the mathematical difficulty of deriving and solving high-dimensional first-order optimality conditions. This paper proposes an efficient least squares Monte Carlo algorithm that is suitable for handling high-dimensional controls. In particular, we first approximate backward recursive dynamic programs on a coarse grid of controls, then use a combined technique of local control regression and adaptive refinement grids to improve the optimal control estimates. We numerically show that the local control regression is more accurate and more efficient than global control regression, and that the overall computational runtime scales polynomially with respect to the dimension of the control. Finally, we further validate our method by solving a portfolio optimization problem with twelve risky assets with transaction cost, liquidity cost and market impact, which involves twelve exogenous risk factors (asset returns) and thirteen endogenous risk factors (portfolio value and asset prices subject to market impact). |
| Date: | 2018–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1803.11467&r=cmp |
| By: | Zuzana Siebertova (Council for Budget Responsibility); Jana Valachyova (Council for Budget Responsibility); Norbert Svarda (Council for Budget Responsibility); Matus Senaj (Council for Budget Responsibility) |
| Abstract: | In this paper, we document the development process of the microsimulation model for the analysis of the indirect value-added tax liabilities of households in Slovakia. This simulation module can be directly integrated into the framework of SIMTASK, the Slovak microsimulation model of income taxes, health and social security contributions and transfers. In the first step, a combined micro-level dataset that integrates information on disposable income and expenditures of Slovak households has been created. Households’ expenditures reported in HBS dataset have been imputed to SK-SILC dataset by estimating parametric Engel curves. Validation of the imputation procedure of households’ consumption and simulation of VAT has been discussed. |
| Keywords: | value added tax, tax and transfer system, income distribution, microsimulation |
| JEL: | C81 D12 D31 H31 |
| Date: | 2018–03 |
| URL: | http://d.repec.org/n?u=RePEc:cbe:dpaper:201801&r=cmp |
| By: | Jared C. Carbone (Division of Economics and Business, Colorado School of Mines); Nicholas Rivers (Graduate School of Public and International Affairs and Institute of the Environment, University of Ottawa); Akio Yamazaki (Department of Economics, University of Calgary); Hidemichi Yonezawa (ETH Zurich) |
| Abstract: | We treat the implementation of the carbon tax in British Columbia as a natural experiment and compare the results of econometric estimates of its effects to counterfactual experiments conducted using an applied general equilibrium (CGE) model of the Canadian economy. The comparison allows us to test the theory-driven predictions of the CGE model. It also allows us to test the identifcation strategy of our econometric model, using the CGE model to indicate under what circumstances general equilibrium policy responses might undermine our attempts at statistical inference. Ex post, we find statistically and economically signifcant effects on sectoral employment levels from the carbon tax --- with levels falling in the most carbon-intensive sectors and rising in the least carbon-intensive. Ex ante, we predict employment responses of very similar sign and magnitude to our econometric measurements (Pearson correlation coefficient of approximately 0.9). We find no evidence to suggest that our difference-in-difference estimator is likely to be undermined by general equilibrium effects in this policy setting. Finally, we explore the use of the econometric estimates to deepen the empirical content of the CGE model. |
| Keywords: | climate policy, carbon tax, computable general equilibrium |
| JEL: | C68 H23 Q54 |
| Date: | 2018–04 |
| URL: | http://d.repec.org/n?u=RePEc:mns:wpaper:wp201802&r=cmp |
| By: | Michael Ludkovski; Aditya Maheshwari |
| Abstract: | We consider solution of stochastic storage problems through regression Monte Carlo (RMC) methods. Taking a statistical learning perspective, we develop the dynamic emulation algorithm (DEA) that unifies the different existing approaches in a single modular template. We then investigate the two central aspects of regression architecture and experimental design that constitute DEA. For the regression piece, we discuss various non-parametric approaches, in particular introducing the use of Gaussian process regression in the context of stochastic storage. For simulation design, we compare the performance of traditional design (grid discretization), against space-filling, and several adaptive alternatives. The overall DEA template is illustrated with multiple examples drawing from natural gas storage valuation and optimal control of back-up generator in a microgrid. |
| Date: | 2018–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1803.11309&r=cmp |
| By: | INOUE Hiroyasu; TODO Yasuyuki |
| Abstract: | This paper reports simulations of supply chain disruptions regarding the Great East Japan Earthquake and the predicted Nankai Trough Earthquake. The simulations are based on the actual nationwide supply chains of Japan and on an agent-based model. As a result, we obtain the following findings. (1) Based on simulations of the Great East Japan Earthquake, we calibrate the parameters in the model. The result shows that the simulation reproduces the aftermath of the disaster well, which means the simulation captures the propagations of the damages and the recoveries on supply chains. (2) Indirect damages of both earthquakes geographically permeate the entire country in quite a short term. Additionally, the damages to firms show synchronized fluctuations due to the network structure. (3) Simulations of the Nankai Trough Earthquake show that direct damages are 12 times greater than those from the Great East Japan Earthquake, but indirect damages are approximately 4.5 times greater in a year. (4) By estimating indirect damage triggered by a single firm loss, approximately 10% of firms cause more than 10% of the damage of the entire supply chains. |
| Date: | 2018–03 |
| URL: | http://d.repec.org/n?u=RePEc:eti:dpaper:18013&r=cmp |
| By: | Ioane Muni Toke (MICS - Mathématiques et Informatique pour la Complexité et les Systèmes - CentraleSupélec, ERIM - Equipe de Recherche en Informatique et Mathématiques - Université de la Nouvelle-Calédonie, CREST - Core Research for Evolutional Science and Technology - JST - Japan Science and Technology Agency); Nakahiro Yoshida (CREST - Core Research for Evolutional Science and Technology - JST - Japan Science and Technology Agency, Graduate School of Mathematical Sciences[Tokyo] - The University of Tokyo) |
| Abstract: | We propose a parametric model for the simulation of limit order books. We assume that limit orders, market orders and cancellations are submitted according to point processes with state-dependent intensities. We propose new functional forms for these intensities, as well as new models for the placement of limit orders and cancellations. For cancellations, we introduce the concept of " priority index " to describe the selection of orders to be cancelled in the order book. Parameters of the model are estimated using likelihood maximization. We illustrate the performance of the model by providing extensive simulation results, with a comparison to empirical data and a standard Poisson reference. |
| Keywords: | order book,limit orders,market orders,cancellations,state-dependent point processes,intensity-based models |
| Date: | 2017–03–06 |
| URL: | http://d.repec.org/n?u=RePEc:hal:journl:hal-01705080&r=cmp |
| By: | Frédéric Abergel (FiQuant - Chaire de finance quantitative - Ecole Centrale Paris); Rémy Tachet Des Combes (FiQuant - Chaire de finance quantitative - Ecole Centrale Paris); Riadh Zaatour (MAS - Mathématiques Appliquées aux Systèmes - EA 4037 - Ecole Centrale Paris) |
| Abstract: | Consistently fitting vanilla option surfaces is an important issue in derivative modelling. In this paper, we consider three different models: local and stochastic volatility, local correlation, hybrid local volatility with stochastic rates, and address their exact, nonparametric calibration. This calibration process requires solving a nonlinear partial integro-differential equation. A modified alternating direction implicit algorithm is used, and its theoretical and numerical analysis is performed. |
| Date: | 2017 |
| URL: | http://d.repec.org/n?u=RePEc:hal:journl:hal-01686114&r=cmp |
| By: | Isil Erel; Léa H. Stern; Chenhao Tan; Michael S. Weisbach |
| Abstract: | Can an algorithm assist firms in their hiring decisions of corporate directors? This paper proposes a method of selecting boards of directors that relies on machine learning. We develop algorithms with the goal of selecting directors that would be preferred by the shareholders of a particular firm. Using shareholder support for individual directors in subsequent elections and firm profitability as performance measures, we construct algorithms to make out-of-sample predictions of these measures of director performance. We then run tests of the quality of these predictions and show that, when compared with a realistic pool of potential candidates, directors predicted to do poorly by our algorithms indeed rank much lower in performance than directors who were predicted to do well. Deviations from the benchmark provided by the algorithms suggest that firm-selected directors are more likely to be male, have previously held more directorships, have fewer qualifications and larger networks. Machine learning holds promise for understanding the process by which existing governance structures are chosen, and has potential to help real world firms improve their governance. |
| JEL: | G34 M12 M51 |
| Date: | 2018–03 |
| URL: | http://d.repec.org/n?u=RePEc:nbr:nberwo:24435&r=cmp |
| By: | Franzen, Stefan (RWTH Aachen University); Madlener, Reinhard (E.ON Energy Research Center, Future Energy Consumer Needs and Behavior (FCN)) |
| Abstract: | This paper presents a real options-based techno-economic analysis of a hydrogen-based wind energy storage system (H2-WESS) deployed adjacent to a nearshore wind farm in northern Germany. The H2-WESS can be used to produce and store hydrogen when feed-in management takes place, in order to avoid the shutdown of wind turbines during times of excess electricity supply, or when the spot market electricity price falls below the estimated (efficiency-adjusted) market price of hydrogen. Moreover, an H2-WESS can provide negative minute reserve capacity. The modular design of the H2-WESS gives an investor the option to expand the capacity and gradually adapt to changing market conditions. The comprehensive and novel simulation model considers all relevant volatile inputs, such as stochastic wind conditions, feed-in management events, prices, and minute reserve calls. By means of a Monte Carlo simulation, annual revenues and their volatility are computed with a view on projected technology improvements until 2030. Based on the simulation results, a binomial real options pricing model is used to design four interdependent binominal trees and to evaluate a Bermuda-type compound expansion option. The decision trees, in which the investor can choose the maximum of the option to either upgrade the H2-WESS to the next expansion stage or to keep the real option alive, feature 390 time steps and 76,050 decision nodes each. Each compound decision takes the option of a smaller expansion stage explicitly into account. The compound expansion option to invest in a 5, 10, 15, or 20 MW H2-WESS has a 15-year expiration time and is found to have a value of about €2 million, compared to the net present value of a 5 MW H-WESS of about €-2.45 million. We conclude from the real options analysis that for a realistic valuation of modular energy projects subject to various uncertainties it is crucial to incorporate the value of managerial flexibility that is influenced. Due to the modular design, and in contrast to conventional power plants, the flexibility of the H2-WESS comprises many specific options. |
| Keywords: | Wind power; Hydrogen; Storage system; Compound expansion option; Monte Carlo simulation; Germany |
| JEL: | Q40 |
| Date: | 2016–06 |
| URL: | http://d.repec.org/n?u=RePEc:ris:fcnwpa:2016_052&r=cmp |
| By: | Michal Horvath (University of York); Matus Senaj (Council for Budget Responsibility); Zuzana Siebertova (Council for Budget Responsibility); Norbert Svarda (Council for Budget Responsibility); Jana Valachyova (Council for Budget Responsibility) |
| Abstract: | The paper introduces a new way of linking microsimulation models with dynamic general equilibrium frameworks to obtain an evaluation of the impact of detailed tax and benefit measures on the aggregate economy. The approach involving polynomial approximation to aggregated output from behavioural microsimulation permits the solution for the long-run steady state and the transition path in one numerical simulation of the dynamic aggregate economy. The practical usefulness of the approach is demonstrated by evaluating actual and hypothetical tax reforms in the context of Slovakia. |
| Keywords: | microsimulation, dynamic general equilibrium, unemployment, labour supply elasticity, tax reform |
| JEL: | E24 H24 H31 J22 |
| Date: | 2018–04 |
| URL: | http://d.repec.org/n?u=RePEc:cbe:wpaper:201801&r=cmp |
| By: | Dhruv Grover (University of California, San Diego); Sebastian Bauhoff (Center for Global Development); Jed Friedman (World Bank) |
| Abstract: | Independent verification is a critical component of performance-based financing (PBF) in health care, in which facilities are offered incentives to increase the volume of specific services but the same incentives may lead them to over-report. We examine alternative strategies for targeted sampling of health clinics for independent verification. Specifically, we empirically compare several methods of random sampling and predictive modeling on data from a Zambian PBF pilot that contains reported and verified performance for quantity indicators of 140 clinics. Our results indicate that machine learning methods, particularly Random Forest, outperform other approaches and can increase the cost-effectiveness of verification activities. |
| Keywords: | performance-based financing, performance verification, audits, machine learning, health care finance, health care providers |
| JEL: | C20 C52 I15 I18 |
| Date: | 2018–04–11 |
| URL: | http://d.repec.org/n?u=RePEc:cgd:wpaper:481&r=cmp |
| By: | Saglam, Ismail |
| Abstract: | In this paper, we provide a welfare ranking for the equilibria of the supply function and quantity competitions in a differentiated product duopoly with demand uncertainty. We prove that the expected consumer surplus is always higher under the supply function competition. By numerical simulations, we also show that if the degree of product substitution is extremely low, then the supply function competition can become a superior form of competition for the duopolistic producers, as well. However, if the degree of product substitution is not extremely low, then the expected producer profits under the supply function competition can be lower than under the quantity competition in situations where the size of the demand uncertainty is below a critical level. We find that this critical level is non-decreasing in the degree of product substitution, while non-increasing both in the marginal cost of producing a unit output and in the own-price sensitivity of each inverse demand curve. Our results imply that in electricity markets with differentiated products, the regulators should not intervene to impose the quantity competition in favor of the supply function competition unless the degree of product substitution is sufficiently high and the predicted demand fluctuations are sufficiently small. |
| Keywords: | Supply function competition; Cournot competition; duopoly; differentiated products; uncertainty |
| JEL: | D43 L13 |
| Date: | 2018 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:85474&r=cmp |
| By: | Giovanni Andrea Cornia |
| Abstract: | The paper examines whether the planned eradication of poverty to the year 2030 part of the Sustainable Development Goals strategy is compatible with the trends expected over the next 15 years in key economic variables such as GDP growth, population growth, income inequality and food prices. To do so, the paper develops a comparative-static, poverty-accounting model that allows to simulate to 2030 the impact on SDG1 (poverty eradication) of the fastest improvements recorded for the above four variables during the last 30 years. Numerous model simulations show that – even under the most favorable assumptions – between 16 and 28 countries (mainly from Africa) out of the 78 analyzed will not reach the SDG1 target. Policy suggestions on how to improve on such results are presented at the end of the paper. |
| Keywords: | SDG1, poverty eradication, inequality, GDP growth, population growth, food prices, public policies. |
| JEL: | D31 I32 J11 Q18 |
| Date: | 2018 |
| URL: | http://d.repec.org/n?u=RePEc:frz:wpaper:wp2018_09.rdf&r=cmp |
| By: | Thomas Gerstner; Bastian Harrach; Daniel Roth |
| Abstract: | The Monte Carlo pathwise sensitivities approach is well established for smooth payoff functions. In this work, we present a new Monte Carlo algorithm that is able to calculate the pathwise sensitivities for discontinuous payoff functions. Our main tool is to combine the one-step survival idea of Glasserman and Staum with the stable differentiation approach of Alm, Harrach, Harrach and Keller. As an application we use the derived results for a two-dimensional calibration of a CoCo-Bond, which we model with different types of discretely monitored barrier options. |
| Date: | 2018–04 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1804.03975&r=cmp |
| By: | Eliaz, Kfir; Spiegler, Ran |
| Abstract: | We study a model in which a "statistician" takes an action on behalf of an agent, based on a random sample involving other people. The statistician follows a penalized regression procedure: the action that he takes is the dependent variable's estimated value given the agent's disclosed personal characteristics. We ask the following question: Is truth-telling an optimal disclosure strategy for the agent, given the statistician's procedure? We discuss possible implications of our exercise for the growing reliance on "machine learning" methods that involve explicit variable selection. |
| Date: | 2018–03 |
| URL: | http://d.repec.org/n?u=RePEc:cpr:ceprdp:12804&r=cmp |