|
on Computational Economics |
|
Issue of 2018‒09‒10
seven papers chosen by |
| By: | Zihao Zhang; Stefan Zohren; Stephen Roberts |
| Abstract: | We develop a large-scale deep learning model to predict price movements from limit order book (LOB) data of cash equities. The architecture utilises convolutional filters to capture the spatial structure of the limit order books as well as LSTM modules to capture longer time dependencies. The model is trained using electronic market quotes from the London Stock Exchange. Our model delivers a remarkably stable out-of-sample prediction accuracy for a variety of instruments and outperforms existing methods such as Support Vector Machines, standard Multilayer Perceptrons, as well as other previously proposed convolutional neural network (CNN) architectures. The results obtained lead to good profits in a simple trading simulation, especially when compared with the baseline models. Importantly, our model translates well to instruments which were not part of the training set, indicating the model's ability to extract universal features. In order to better understand these features and to go beyond a "black box" model, we perform a sensitivity analysis to understand the rationale behind the model predictions and reveal the components of LOBs that are most relevant. The ability to extract robust features which translate well to other instruments is an important property of our model which has many other applications. |
| Date: | 2018–08 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1808.03668&r=cmp |
| By: | Leander L\"ow; Martin Spindler; Eike Brechmann |
| Abstract: | Insurance companies must manage millions of claims per year. While most of these claims are non-fraudulent, fraud detection is core for insurance companies. The ultimate goal is a predictive model to single out the fraudulent claims and pay out the non-fraudulent ones immediately. Modern machine learning methods are well suited for this kind of problem. Health care claims often have a data structure that is hierarchical and of variable length. We propose one model based on piecewise feed forward neural networks (deep learning) and another model based on self-attention neural networks for the task of claim management. We show that the proposed methods outperform bag-of-words based models, hand designed features, and models based on convolutional neural networks, on a data set of two million health care claims. The proposed self-attention method performs the best. |
| Date: | 2018–08 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1808.10543&r=cmp |
| By: | Alfonso Herrera García (Instituto Mexicano del Transporte) |
| Abstract: | This paper explores several methods for coping with excess demand at airports through applying simulation modeling that focusses on how to use the existing airport infrastructure more efficiently. The introduction presents an overview of the importance of solving the airport saturation problem and sets out several approaches to solutions, which are divided into four distinct groups, or options. The fourth option applies operational practices and/or new technology to improve the airport procedures, including computer modeling and simulation. The document presents the application of simulation models to the capacity issues at the Mexico City Airport to demonstrate how to potentially alleviate congestion. Examples include redistribution of takeoffs and landings to increase runway capacity; reduction of air traffic movements through allowing operations of aircraft with greater capacity; deployment of new technologies to increase runway capacity; and by means of new operational procedures, changing the aircraft waiting sequence to reduce delays. |
| Date: | 2017–09–12 |
| URL: | http://d.repec.org/n?u=RePEc:oec:itfaab:2017/28-en&r=cmp |
| By: | Bayer, Christian; Luetticke, Ralph |
| Abstract: | This paper describes a method for solving heterogeneous agent models with aggregate risk and many idiosyncratic states formulated in discrete time. It extends the method proposed by Reiter (2009) and complements recent work by Ahn et al. (2017) on how to solve such models in continuous time. We suggest first solving for the stationary equilibrium of the model without aggregate risk. We then write the functionals that describe the recursive equilibrium as sparse expansions around their stationary equilibrium counterparts. Finally we use the perturbation method of Schmitt-Grohé and Uribe (2004) to approximate the aggregate dynamics of the model. |
| Keywords: | Heterogeneous Agent Models; incomplete markets; linearization; Numerical Methods |
| JEL: | C63 E32 |
| Date: | 2018–07 |
| URL: | http://d.repec.org/n?u=RePEc:cpr:ceprdp:13071&r=cmp |
| By: | Christian Bayer; Martin Redmann; John Schoenmakers |
| Abstract: | We introduce new variants of classical regression-based algorithms for optimal stopping problems based on computation of regression coefficients by Monte Carlo approximation of the corresponding $L^2$ inner products instead of the least-squares error functional. Coupled with new proposals for simulation of the underlying samples, we call the approach "pseudo-regression". We show that the approach leads to asymptotically smaller errors, as well as less computational cost. The analysis is justified by numerical examples. |
| Date: | 2018–08 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1808.04725&r=cmp |
| By: | Hajibabaei, Ehsan; Ghasmi, Alireza; Hosseini, Seyed Abbas |
| Abstract: | Non-conservative dense flow frontal velocity has been simulated two dimensionally by fluent numerical code. The outcomes have been compared with experimental results. Numerical simulation was conducted as two-phase through Euler-Lagrange method. Reynolds-Stress Turbulent Model (RSM) with non-uniform grid and shredding mesh on the channel floor. The results obtained from numerical model of head frontal velocity show a good compliance with experiment results and greatly help analyzing the pattern of fluid movement in different scales. |
| Keywords: | dense flow, fluent, head frontal velocity |
| JEL: | L63 L65 Q1 Q15 Q16 Q51 Q53 Q55 |
| Date: | 2017–02 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:88383&r=cmp |
| By: | Turkmen Goksel (Ankara University) |
| Abstract: | A standard model of international trade has a setting with constant labor supply. However, this model introduces a consumption-leisure choice into a traditional model of international economics. Therefore, this paper focuses on the response of labor to changes in tariffs. Moreover, I show that there exists a positive optimal tariff rate which maximizes welfare in a setting with endogenous labor and compare this result quantitatively with the standard models using constant labor supply. This paper also focuses on the welfare implications of a decline in trade barriers (in terms of tariffs). I utilize a version of computational general equilibrium model of international trade (based on Armington assumption) where countries are potentially asymmetric in terms of labor endowment, productivity, etc. Eaton and Kortum (2002) derive a simple formula which shows the gains from trade and this formula is generalized by Arkolakis, Costinot, and Rodriguez-Clare (2012) in the case of iceberg costs and exogenously fixed labor supply. I generalize this formula in Armington setup with tariffs and endogenous labor supply and highlight the importance of both revenue generating tariffs and consumption-leisure choice. |
| Keywords: | Endogenous Labor Supply, Optimal Tariff, Computational General Equilibrium, Welfare |
| JEL: | F10 F11 F16 |
| Date: | 2018–07 |
| URL: | http://d.repec.org/n?u=RePEc:sek:iacpro:7809653&r=cmp |