nep-cmp New Economics Papers
on Computational Economics
Issue of 2018‒10‒22
thirteen papers chosen by
Stan Miles
Thompson Rivers University

  1. A Thick ANN Model for Forecasting Inflation By Muhammad Nadim Hanif; Khurrum S. Mughal; Javed Iqbal
  2. DEEP LEARNING NEURAL NETWORKS AS A MODEL OF SACCADIC GENERATION By Sofia Krasovskaya; Georgiy Zhulikov; Joseph MacInnes
  3. Early Detection of Students at Risk - Predicting Student Dropouts Using Administrative Student Data and Machine Learning Methods By Schneider, Kerstin; Berens, Johannes; Oster, Simon; Burghoff, Julian
  4. Public Procurement and Reputation: An Agent-Based Model By Nadia Fiorino; Emma Galli; Ilde Rizzo; Marco Valente
  5. Efficient simulation of clustering jumps with CIR intensity By Dassios, Angelos; Zhao, Hongbiao
  6. Exploring the General Equilibrium Costs of Sector-Specific Environmental Regulations By Alex L. Marten; Richard Garbaccio; Ann Wolverton
  7. An Efficient Approach for Removing Look-ahead Bias in the Least Square Monte Carlo Algorithm: Leave-One-Out By Jaehyuk Choi; Chenru Liu; Jeechul Woo
  8. Housing Tax Policy: Comment By Hamed Ghiaie; Jean-François Rouillard
  9. Mean-Field Games with Differing Beliefs for Algorithmic Trading By Philippe Casgrain; Sebastian Jaimungal
  10. Modelling Competitive Mortgage Termination Option Strategies: Default vs Restructuring and Prepayment vs Defeasance By Lok Man Michel Tong; Gianluca Marcato
  11. Identifying Influential Traders by Agent Based Modelling By Kopp, Thomas; Salecker, Jan
  12. A Complex Model of Consumer Food Acquisitions: Applying Machine Learning and Directed Acyclic Graphs to the National Household Food Acquisition and Purchase Survey (FoodAPS) By Senia, Mark C.; Dharmasena, Senarath; Todd, Jessica E.
  13. Adopting Bio-Energy Crops: Does Farmers’ Attitude toward Loss Matter? By Anand, Mohit; Miao, Ruiqing; Khanna, Madhu

  1. By: Muhammad Nadim Hanif (State Bank of Pakistan); Khurrum S. Mughal (State Bank of Pakistan); Javed Iqbal (State Bank of Pakistan)
    Abstract: Inflation forecasting is an essential activity at central banks to formulate forward looking monetary policy stance. Like in other fields, machine learning is finding its way to forecasting; inflation forecasting is not any exception. In machine learning, most popular tool for forecasting is artificial neural network (ANN). Researchers have used different performance measures (including RMSE) to optimize set of characteristics - architecture, training algorithm and activation function - of an ANN model. However, any chosen ‘optimal’ set may not remain reliable on realization of new data. We suggest use of ‘mode’ or most appearing set from a simulation based distribution of optimum ‘set of characteristics of ANN model’; selected from a large number of different sets. Here again, we may have a different trained network in case we re-run this ‘modal’ optimal set since initial weights in training process are assigned randomly. To overcome this issue, we suggest use of ‘thickness’ to produce stable and reliable forecasts using modal optimal set. Using January 1958 to December 2017 year on year (YoY) inflation data of Pakistan, we found that our YoY inflation forecasts (based on aforementioned multistage forecasting scheme) outperform those from a number of inflation forecasting models of Pakistan economy.
    Keywords: Artificial Neural Networks, Inflation Forecasting
    JEL: C45 E31 E37
    Date: 2018–10
  2. By: Sofia Krasovskaya (National Research University Higher School of Economics); Georgiy Zhulikov (National Research University Higher School of Economics); Joseph MacInnes (National Research University Higher School of Economics)
    Abstract: Approximately twenty years ago, Laurent Itti and Christof Koch created a model of saliency in visual attention in an attempt to recreate the work of biological pyramidal neurons by mimicking neurons with centre-surround receptive fields. The Saliency Model has launched many studies that contributed to the understanding of layers of vision and the sphere of visual attention. The aim of the current study is to improve this model by using an artificial neural network that generates saccades similar to how humans make saccadic eye movements. The proposed model uses a Leaky Integrate-and-Fire layer for temporal predictions, and replaces parallel feature maps with a deep learning neural network in order to create a generative model that is precise for both spatial and temporal predictions. Our deep neural network was able to predict eye movements based on unsupervised learning from raw image input, as well as supervised learning from fixation maps retrieved during an eye-tracking experiment conducted with 35 participants at later stages in order to train a 2D softmax layer. The results imply that it is possible to match the spatial and temporal distributions of the model to spatial and temporal human distributions.
    Keywords: saccade generation, salience model, deep learning neural network, visual search, leaky integrate and fire
    JEL: Z
    Date: 2018
  3. By: Schneider, Kerstin; Berens, Johannes; Oster, Simon; Burghoff, Julian
    Abstract: High rates of student attrition in tertiary education are a major concern for universities and public policy, as dropout is not only costly for the students but also wastes public funds. To successfully reduce student attrition, it is imperative to understand which students are at risk of dropping out and what are the underlying determinants of dropout. We develop an early detection system (EDS) that uses machine learning and classic regression techniques to predict student success in tertiary education as a basis for a targeted intervention. The method developed in this paper is highly standardized and can be easily implemented in every German institution of higher education, as it uses student performance and demographic data collected, stored, and maintained by legal mandate at all German universities and therefore self-adjusts to the university where it is employed. The EDS uses regression analysis and machine learning methods, such as neural networks, decision trees and the AdaBoost algorithm to identify student characteristics which distinguish potential dropouts from graduates. The EDS we present is tested and applied on a medium-sized state university with 23,000 students and a medium-sized private university of applied sciences with 6,700 students. Our results indicate a prediction accuracy at the end of the 1st semester of 79% for the state university and 85% for the private university of applied sciences. Furthermore, accuracy of the EDS increases with each completed semester as new performance data becomes available. After the fourth semester, the accuracy improves to 90% for the state university and 95% for the private university of applied sciences.
    Keywords: student attrition,early detection,administrative data,higher education,machine learning,AdaBoost
    JEL: I23 C45 H52
    Date: 2018
  4. By: Nadia Fiorino (University of L'Aquila, Italy); Emma Galli (University of Rome "Sapienza", Italy); Ilde Rizzo (University of Catania, Italy); Marco Valente (University of L’Aquila; LEM Sant’Anna, Pisa (Italy); SPRU, University of Sussex (UK) and Ruhr-Universit¨at Bochum (Germany))
    Abstract: Based on the literature on public procurement regulation, we use an Agent-Based Model to assess the performance of different selection procedures. Specifically, we aim at investigating whether and how the inclusion of reputation of firms in the public procurement selection process affects the final cost of the contract. The model defines two types of actors: i) firms potentially competing to win the contract; ii) a contracting authority, aiming at minimizing procurement costs. These actors respond to environmental conditions affecting the actual costs of carrying on the project and which are unknown to firms and to the contracting authority at the time of bidding. The results from the model are generated through simulations by considering different configurations and varying some parameters of the model, such as the firms’ skills, the level of opportunistic rebate, the relative weight of reputation and rebate. The main conclusion is that reputation matters and some policy implications are drawn.
    Keywords: Public works, Procurement, Agent-Based modelling JEL Classification code: H57, L14, C63
    Date: 2018–10
  5. By: Dassios, Angelos; Zhao, Hongbiao
    Abstract: We introduce a broad family of generalised self-exciting point processes with CIR-type intensities, and develop associated algorithms for their exact simulation. The underlying models are extensions of the classical Hawkes process, which already has numerous applications in modelling the arrival of events with clustering or contagion effect in finance, economics and many other fields. Interestingly, we find that the CIR-type intensity together with its point process can be sequentially decomposed into simple random variables, which immediately leads to a very efficient simulation scheme. Our algorithms are also pretty accurate and flexible. They can be easily extended to further incorporate externally-excited jumps, or, to a multidimensional framework. Some typical numerical examples and comparisons with other well known schemes are reported in detail. In addition, a simple application for modelling a portfolio loss process is presented.
    Keywords: contagion risk; jump clustering; stochastic intensity model; self-exciting point process; self-exciting point process with CIR intensity; Hawkes process; CIR process; square-root process; exact simulation; Monte Carlo simulation; portfolio risk
    JEL: C15 C53 C63
    Date: 2017–11–01
  6. By: Alex L. Marten; Richard Garbaccio; Ann Wolverton
    Abstract: The requisite scope of analysis to adequately estimate the social cost of environmental regulations has been subject to much discussion. The literature has demonstrated that engineering or partial equilibrium cost estimates likely underestimate the social cost of large-scale environmental regulations and environmental taxes. However, the conditions under which general equilibrium (GE) analysis adds value to welfare analysis for single-sector technology or performance standards, the predominant policy intervention in practice, remains an open question. Using a numerical computable general equilibrium (CGE) model, we investigate the GE effects of regulations across different sectors, abatement technologies, and regulatory designs. Our results show that even for small regulations the GE effects are significant, and that engineering estimates of compliance costs can substantially underestimate the social cost of single-sector environmental regulations. We find the downward bias from using engineering costs to approximate social costs depends on the input composition of abatement technologies and the regulated sector.
    Keywords: environmental regulation, general equilibrium, social costs
    JEL: D58 Q52 Q58
    Date: 2018–10
  7. By: Jaehyuk Choi; Chenru Liu; Jeechul Woo
    Abstract: The least square Monte Carlo (LSM) algorithm proposed by Longstaff and Schwartz [2001] is the most widely used method for pricing options with early exercise features. The LSM estimator contains look-ahead bias, and the conventional technique of removing it necessitates an independent set of simulations. This study proposes a new approach for efficiently eliminating look-ahead bias by using the leave-one-out method, a well-known cross-validation technique for machine learning applications. The leave-one-out LSM (LOOLSM) method is illustrated with examples, including multi-asset options whose LSM price is biased high. The asymptotic behavior of look-ahead bias is also discussed with the LOOLSM approach.
    Date: 2018–10
  8. By: Hamed Ghiaie (Département d'économique, Université de Cergy-Pontoise); Jean-François Rouillard (Département d'économique, Université de Sherbrooke)
    Abstract: Alpanda and Zubairy (2016) examine the effects of permanent changes to four types of housing-related tax policies in the context of a multi-agent DGE model. They find long-run tax multipliers that range from -2.21 to -1.53. However, we find an error in their codes that has a significant impact on the size of these multipliers. We correct their error and re-simulate their model. The long-run multipliers we find are reduced almost in half—they now range from -1.25 to -0.84. We also com- pute short-run multipliers at a 20-quarter horizon and find much lower multipliers, ranging between -0.14 to -0.02.
    Keywords: Housing taxation, banking, dynamic general equilibrium.
    JEL: E62 G28 H24 R38
    Date: 2018–10
  9. By: Philippe Casgrain; Sebastian Jaimungal
    Abstract: Even when confronted with the same data, agents often disagree on a model of the real-world. Here, we address the question of how interacting heterogenous agents, who disagree on what model the real-world follows, optimize their trading actions. The market has latent factors that drive prices, and agents account for the permanent impact they have on prices. This leads to a large stochastic game, where each agents' performance criteria is computed under a different probability measure. We analyse the mean-field game (MFG) limit of the stochastic game and show that the Nash equilibria is given by the solution to a non-standard vector-valued forward-backward stochastic differential equation. Under some mild assumptions, we construct the solution in terms of expectations of the filtered states. We prove the MFG strategy forms an \epsilon-Nash equilibrium for the finite player game. Lastly, we present a least-squares Monte Carlo based algorithm for computing the optimal control and illustrate the results through simulation in market where agents disagree on the model.
    Date: 2018–10
  10. By: Lok Man Michel Tong; Gianluca Marcato
    Abstract: We build a two-stage competing risk model for pricing four types of early Termination options written on commercial mortgages: default vs restructuring and prepayment vs defeasance as two pairs of competitions. It is the first study to consider restructuring as a “competitor” with default. The key feature of our model is to introduce collateral underlying property market supply constraints into a property Price process which would determine values of early termination options. Our simulations find out greater probability to restructure mortgages by reducing interest and extending maturity and to prepay in cash. We also prove that tightening property supply constraints pushes up values of default, restructuring and prepayment by pricing their analogous options: default (a series of compound European Call on Put options), mortgage restructuring (an exchange option between mortgages with different cash flow structures), prepayment in cash (a series of compound European Call on Call options),and defeasance (an exchange option of more liquid assets with less liquid ones)in different scenarios. Therefore, we suggest controlling property supply constraints as an alternative risk management measure for mortgage markets.
    Keywords: Defeasance; Mortgage Default; Prepayment; Property Supply Constraints; Restructuring
    JEL: R3
    Date: 2018–01–01
  11. By: Kopp, Thomas; Salecker, Jan
    Keywords: Ag Finance and Farm Management, Behavioral & Institutional Economics, Demand and Price Analysis
    Date: 2018–06–20
  12. By: Senia, Mark C.; Dharmasena, Senarath; Todd, Jessica E.
    Abstract: Complex causal relationships among a large set of variables that affect the U.S. households’ food acquisition and purchase decisions were estimated using machine learning algorithms and directed acyclic graphs. Asians and Hispanics live in an environment with high concentrations of fast- and non-fast food restaurants. Obesity is less prevalent among Asians. Being Hispanic makes one to be more food insecure. Those with higher incomes are food secure and obesity is less prevalent among them. Being Black positively causes to be a SNAP participant and food insecure. Obesity is positively caused by fair/poor health and diet status.
    Keywords: Consumer/Household Economics, Food Consumption/Nutrition/Food Safety
    Date: 2018–01–15
  13. By: Anand, Mohit; Miao, Ruiqing; Khanna, Madhu
    Abstract: This paper investigates farmers’ willingness to grow bio-energy crops (namely, miscanthus and switchgrass) while accounting for their preferences toward loss. We model a representative farmer’s optimal land allocation problem between conventional crops and bio-energy crops by employing the prospect theory. Numerical simulation is conducted for 1,919 U.S. counties east of the 100th Meridian that have yield data for corn and for at least one bio-energy crop. Results show that all else equal, if farmers are credit constrained then accounting for loss aversion will decrease the miscanthus production but increase switchgrass production. If farmers are not credit constrained, however, then accounting for loss aversion only has small impact on bio-energy crop production, indicating that the availability of credit mitigates the effect of farmers’ loss preferences. We also find that biomass production on marginal land is less sensitive to farmers’ loss aversion than production on high quality land is, which underscores the importance of marginal land in providing biomass for the bio-energy and bio-product sector. Moreover, results show that impact of loss aversion is smaller when interest rate is low as compare to scenarios under which interest rate is high. Geographical configuration of biomass production under various loss aversion, credit constraint, and interest rate scenarios are examined as well.
    Keywords: Production Economics, Risk and Uncertainty
    Date: 2018–01–04

This nep-cmp issue is ©2018 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.