|
on Computational Economics |
|
Issue of 2020‒09‒21
twenty papers chosen by |
| By: | Shuaiqiang Liu; Lech A. Grzelak; Cornelis W. Oosterlee |
| Abstract: | We propose an accurate data-driven numerical scheme to solve Stochastic Differential Equations (SDEs), by taking large time steps. The SDE discretization is built up by means of a polynomial chaos expansion method, on the basis of accurately determined stochastic collocation (SC) points. By employing an artificial neural network to learn these SC points, we can perform Monte Carlo simulations with large time steps. Error analysis confirms that this data-driven scheme results in accurate SDE solutions in the sense of strong convergence, provided the learning methodology is robust and accurate. With a variant method called the compression-decompression collocation and interpolation technique, we can drastically reduce the number of neural network functions that have to be learned, so that computational speed is enhanced. Numerical results shows the high quality strong convergence error results, when using large time steps, and the novel scheme outperforms some classical numerical SDE discretizations. Some applications, here in financial option valuation, are also presented. |
| Date: | 2020–09 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2009.03202&r=all |
| By: | Michael Rollins; Dave Cliff |
| Abstract: | This paper presents novel results generated from a new simulation model of a contemporary financial market, that cast serious doubt on the previously widely accepted view of the relative performance of various well-known public-domain automated-trading algorithms. Various public-domain trading algorithms have been proposed over the past 25 years in a kind of arms-race, where each new trading algorithm was compared to the previous best, thereby establishing a "pecking order", i.e. a partially-ordered dominance hierarchy from best to worst of the various trading algorithms. Many of these algorithms were developed and tested using simple minimal simulations of financial markets that only weakly approximated the fact that real markets involve many different trading systems operating asynchronously and in parallel. In this paper we use BSE, a public-domain market simulator, to run a set of experiments generating benchmark results from several well-known trading algorithms. BSE incorporates a very simple time-sliced approach to simulating parallelism, which has obvious known weaknesses. We then alter and extend BSE to make it threaded, so that different trader algorithms operate asynchronously and in parallel: we call this simulator Threaded-BSE (TBSE). We then re-run the trader experiments on TBSE and compare the TBSE results to our earlier benchmark results from BSE. Our comparison shows that the dominance hierarchy in our more realistic experiments is different from the one given by the original simple simulator. We conclude that simulated parallelism matters a lot, and that earlier results from simple simulations comparing different trader algorithms are no longer to be entirely trusted. |
| Date: | 2020–09 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2009.06905&r=all |
| By: | Pihnastyi, Oleh; Khodusov, Valery |
| Abstract: | In this paper, a model of a transport conveyor system using a neural network is demonstrated. The analysis of the main parameters of modern conveyor systems is presented. The main models of the conveyor section, which are used for the design of control systems for flow parameters, are considered. The necessity of using neural networks in the design of conveyor transport control systems is substantiated. A review of conveyor models using a neural network is performed. The conditions of applicability of models using neural networks to describe conveyor systems are determined. A comparative analysis of the analytical model of the conveyor section and the model using the neural network is performed. The technique of forming a set of test data for the process of training a neural network is presented. The foundation for the formation of test data for learning neural network is an analytical model of the conveyor section. Using an analytical model allowed us to form a set of test data for transient dynamic modes of functioning of the transport system. The transport system is presented in the form of a directed graph without cycles. Analysis of the model using a neural network showed a high-quality relationship between the output flow for different conveyor sections of the transport system |
| Keywords: | conveyor; PDE– model; distributed system; transport delay |
| JEL: | C02 C15 C25 C44 D24 |
| Date: | 2020–05–01 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:101527&r=all |
| By: | Zhengxin Joseph Ye; Bjorn W. Schuller |
| Abstract: | While Post-Earnings-Announcement Drift (PEAD) is one of the most studied stock market anomalies, the current literature is often limited in explaining this phenomenon by a small number of factors using simpler regression methods. In this paper, we use a machine learning based approach instead, and aim to capture the PEAD dynamics using data from a large group of stocks and a wide range of both fundamental and technical factors. Our model is built around the Extreme Gradient Boosting (XGBoost) and uses a long list of engineered input features based on quarterly financial announcement data from 1,106 companies in the Russell 1000 index between 1997 and 2018. We perform numerous experiments on PEAD predictions and analysis and have the following contributions to the literature. First, we show how Post-Earnings-Announcement Drift can be analysed using machine learning methods and demonstrate such methods' prowess in producing credible forecasting on the drift direction. It is the first time PEAD dynamics are studied using XGBoost. We show that the drift direction is in fact driven by different factors for stocks from different industrial sectors and in different quarters and XGBoost is effective in understanding the changing drivers. Second, we show that an XGBoost well optimised by a Genetic Algorithm can help allocate out-of-sample stocks to form portfolios with higher positive returns to long and portfolios with lower negative returns to short, a finding that could be adopted in the process of developing market neutral strategies. Third, we show how theoretical event-driven stock strategies have to grapple with ever changing market prices in reality, reducing their effectiveness. We present a tactic to remedy the difficulty of buying into a moving market when dealing with PEAD signals. |
| Date: | 2020–09 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2009.03094&r=all |
| By: | Qiao Zhou; Ningning Liu |
| Abstract: | The prediction of a stock price has always been a challenging issue, as its volatility can be affected by many factors such as national policies, company financial reports, industry performance, and investor sentiment etc.. In this paper, we present a prediction model based on deep CNN and the candle charts, the continuous time stock information is processed. According to different information richness, prediction time interval and classification method, the original data is divided into multiple categories as the training set of CNN. In addition, the convolutional neural network is used to predict the stock market and analyze the difference in accuracy under different classification methods. The results show that the method has the best performance when the forecast time interval is 20 days. Moreover, the Moving Average Convergence Divergence and three kinds of moving average are added as input. This method can accurately predict the stock trend of the US NDAQ exchange for 92.2%. Meanwhile, this article distinguishes three conventional classification methods to provide guidance for future research. |
| Date: | 2020–09 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2009.03239&r=all |
| By: | Yan Wang; Xuelei Sherry Ni |
| Abstract: | In the peer-to-peer (P2P) lending market, lenders lend the money to the borrowers through a virtual platform and earn the possible profit generated by the interest rate. From the perspective of lenders, they want to maximize the profit while minimizing the risk. Therefore, many studies have used machine learning algorithms to help the lenders identify the "best" loans for making investments. The studies have mainly focused on two categories to guide the lenders' investments: one aims at minimizing the risk of investment (i.e., the credit scoring perspective) while the other aims at maximizing the profit (i.e., the profit scoring perspective). However, they have all focused on one category only and there is seldom research trying to integrate the two categories together. Motivated by this, we propose a two-stage framework that incorporates the credit information into a profit scoring modeling. We conducted the empirical experiment on a real-world P2P lending data from the US P2P market and used the Light Gradient Boosting Machine (lightGBM) algorithm in the two-stage framework. Results show that the proposed two-stage method could identify more profitable loans and thereby provide better investment guidance to the investors compared to the existing one-stage profit scoring alone approach. Therefore, the proposed framework serves as an innovative perspective for making investment decisions in P2P lending. |
| Date: | 2020–09 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2009.04536&r=all |
| By: | Benjamin Patrick Evans; Kirill Glavatskiy; Michael S. Harr\'e; Mikhail Prokopenko |
| Abstract: | Housing markets are inherently spatial, yet many existing models fail to capture this spatial dimension. Here we introduce a new graph-based approach for incorporating a spatial component in a large-scale urban housing agent-based model (ABM). The model explicitly captures several social and economic factors that influence the agents' decision-making behaviour (such as fear of missing out, their trend following aptitude, and the strength of their submarket outreach), and interprets these factors in spatial terms. The proposed model is calibrated and validated with the housing market data for the Greater Sydney region. The ABM simulation results not only include predictions for the overall market, but also produce area-specific forecasting at the level of local government areas within Sydney. In addition, the simulation results elucidate movement patterns across submarkets, in both spatial and homeownership terms, including renters, first-time home buyers, as well as local and overseas investors. |
| Date: | 2020–09 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2009.06914&r=all |
| By: | Calvin Price (MUFG Bank) |
| Abstract: | Capturing interdependencies among many variables is a crucial part of economic forecasting. We show how multiple estimated equations can be solved simultaneously with the Stata forecast command and how to simulate the system through time to produce forecasts. This can be combined with user-defined exogenous variables, so that different assumptions can be used to create forecasts under different scenarios. Techniques for assessing the quality of both ex post and ex ante forecasts are shown, along with a simple example model of the U.S. economy. |
| Date: | 2020–08–20 |
| URL: | http://d.repec.org/n?u=RePEc:boc:scon20:5&r=all |
| By: | Eric Benhamou; David Saltiel; Jean-Jacques Ohana; Jamal Atif |
| Abstract: | Deep reinforcement learning (DRL) has reached super human levels in complex tasks like game solving (Go and autonomous driving). However, it remains an open question whether DRL can reach human level in applications to financial problems and in particular in detecting pattern crisis and consequently dis-investing. In this paper, we present an innovative DRL framework consisting in two sub-networks fed respectively with portfolio strategies past performances and standard deviations as well as additional contextual features. The second sub network plays an important role as it captures dependencies with common financial indicators features like risk aversion, economic surprise index and correlations between assets that allows taking into account context based information. We compare different network architectures either using layers of convolutions to reduce network's complexity or LSTM block to capture time dependency and whether previous allocations is important in the modeling. We also use adversarial training to make the final model more robust. Results on test set show this approach substantially over-performs traditional portfolio optimization methods like Markowitz and is able to detect and anticipate crisis like the current Covid one. |
| Date: | 2020–09 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2009.07200&r=all |
| By: | Joop van de Heijning; Stephan Leitner; Alexandra Rausch |
| Abstract: | This work researches the impact of including a wider range of participants in the strategy-making process on the performance of organizations which operate in either moderately or highly complex environments. Agent-based simulation demonstrates that the increased number of ideas generated from larger and diverse crowds and subsequent preference aggregation lead to rapid discovery of higher peaks in the organization's performance landscape. However, this is not the case when the expansion in the number of participants is small. The results confirm the most frequently mentioned benefit in the Open Strategy literature: the discovery of better performing strategies. |
| Date: | 2020–09 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2009.04912&r=all |
| By: | Song, Suihong; Mukerji, Tapan; Hou, Jiagen |
| Abstract: | Conditional facies modeling combines geological spatial patterns with different types of observed data, to build earth models for predictions of subsurface resources. Recently, researchers have used generative adversarial networks (GANs) for conditional facies modeling, where an unconditional GAN is first trained to learn the geological patterns using the original GANs loss function, then appropriate latent vectors are searched to generate facies models that are consistent with the observed conditioning data. A problem with this approach is that the time-consuming search process needs to be conducted for every new conditioning data. As an alternative, we improve GANs for conditional facies modeling by introducing an extra condition-based loss function and adjusting the architecture of the generator to take the conditioning data as inputs, based on progressive growing of GANs. The condition-based loss function is defined as the inconsistency between the input conditioning value and the corresponding characteristics exhibited by the output facies model, and forces the generator to learn the ability of being consistent with the input conditioning data, together with the learning of geological patterns. Our input conditioning factors include global features (e.g. the mud facies proportion) alone, local features such as sparse well facies data alone, and joint combination of global features and well facies data. After training, we evaluate both the quality of generated facies models and the conditioning ability of the generators, by manual inspection and quantitative assessment. The trained generators are quite robust in generating high-quality facies models conditioned to various types of input conditioning information. |
| Date: | 2020–06–29 |
| URL: | http://d.repec.org/n?u=RePEc:osf:eartha:fm24b&r=all |
| By: | Asim Kumer Dey; Toufiqul Haq; Kumer Das; Yulia R. Gel |
| Abstract: | We investigate the impact of Covid-19 cases and deaths, local spread spreads of Covid-19, and Google search activities on the US stock market. We develop a temporal complex network to quantify US county level spread dynamics of Covid-19. We conduct the analysis by using the following sequence of methods: Spearman's rank correlation, Granger causality, Random Forest (RF) model, and EGARCH (1,1) model. The results suggest that Covid-19 cases and deaths, its local spread spreads, and Google searches have impacts on the abnormal stock price between January 2020 to May 2020. However, although a few of Covid-19 variables, e.g., US total deaths and US new cases exhibit causal relationship on price volatility, EGARCH model suggests that Covid-19 cases and deaths, local spread spreads of Covid-19, and Google search activities do not have impacts on price volatility. |
| Date: | 2020–08 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2008.10885&r=all |
| By: | Castiel, Eyal; Borst, Sem; Miclo, Laurent; Simatos, Florian; Whiting, Phil |
| Abstract: | We examine a queue-based random-access algorithm where activation and deactivation rates are adapted as functions of queue lengths. We establish its heavy traffic behavior on a complete interference graph, which turns out to be nonstandard in two respects: (1) the scaling depends on some parameter of the algorithm and is not the N/N2 scaling usually found in functional central limit theorems; (2) the heavy traffic limit is deterministic. We discuss how this nonstandard behavior arises from the idleness induced by the distributed nature of the algorithm. In order to prove our main result, we develop a new method for obtaining a fully coupled stochastic averaging principle. |
| Date: | 2020–08 |
| URL: | http://d.repec.org/n?u=RePEc:tse:wpaper:124587&r=all |
| By: | Patrick Reinwald; Stephan Leitner; Friederike Wall |
| Abstract: | We follow the agentization approach and transform the standard-hidden action model introduced by Holmstr\"om into an agent-based model. Doing so allows us to relax some of the incorporated rather "heroic" assumptions related to the (i) availability of information about the environment and the (ii) principal's and agent's cognitive capabilities (with a particular focus on their memory). In contrast to the standard hidden-action model, the principal and the agent are modeled to learn about the environment over time with varying capabilities to process the learned pieces of information. Moreover, we consider different characteristics of the environment. Our analysis focuses on how close and how fast the incentive scheme, which endogenously emerges from the agent-based model, converges to the second-best solution proposed by the standard hidden-action model. Also, we investigate whether a stable solution can emerge from the agent-based model variant. The results show that in stable environments the emergent result can nearly reach the solution proposed by the standard hidden-action model. Surprisingly, the results indicate that turbulence in the environment leads to stability in earlier time periods. |
| Date: | 2020–09 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2009.07124&r=all |
| By: | Billy Buchanan (Fayette County Public Schools) |
| Abstract: | Have you ever created a program that requires a nontrivial amount of data to be present or available (for example, look-up/value tables, data used for the program interface, etc…)? If you have, you’ll likely have experienced the performance penalty that multiple I/O operations can cause. In this talk, I’ll provide an example of how to implement a common programming pattern from the computer science field and how it can solve this performance issue more effectively. Based on a set of scripts developed by Adam Nelson (https://github.com/adamrossnelson/StataIPEDSAll), I developed a solution (https://github.com/wbuchanan/ipeds) that uses the singleton pattern to reduce object instantiation and I/O operations over multiple calls in order to improve performance. |
| Date: | 2020–08–20 |
| URL: | http://d.repec.org/n?u=RePEc:boc:scon20:9&r=all |
| By: | Foltas, Alexander |
| Abstract: | I use textual data to model German professional macroeconomic forecasters' information sets and use machine-learning techniques to analyze the efficiency of forecasts. To this end, I extract information from forecast reports using a combination of topic models and word embeddings. I then use this information and traditional macroeconomic predictors to study the efficiency of investment forecasts. |
| Keywords: | Forecast Efficiency,Investment,Random Forest,Topic Modeling |
| JEL: | C53 E27 E22 |
| Date: | 2020 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:pp1859:19&r=all |
| By: | RODOMANOV Anton, (Université catholique de Louvain, CORE, Belgium); NESTEROV Yurii, (Université catholique de Louvain, CORE, Belgium) |
| Abstract: | In this paper, we study greedy variants of quasi-Newton methods. They are based on the updating forulas from a certain subclass of the Broyden family. In particular, this subclass includes the well-known DFP, BFGS ans SR1 updates. However, in contrast to the classical quasi-Newton methods, which use the difference of successive iterates for updating the Hessian approximations, our methods apply basis vectors, greedily selected so as to maximize a certain measure of progress. For greedy quasi-Newton methods, we estabish an explicit non-asymptotic bound on their rate of local superlinear convergence, which contains a contracting factor, depending on the square of the iteration counter. We also show that these methods produce Hessian approximations whose deviation from the exact Hessians linearly convergences to zero. |
| Keywords: | quasi-Newton methods, Broyden family, SR1, DFP, BFGS, superlinear convergence, local convergence, rate of convergence |
| JEL: | F12 R12 |
| Date: | 2020–02–10 |
| URL: | http://d.repec.org/n?u=RePEc:cor:louvco:2020006&r=all |
| By: | Matthias Schonlau (University of Waterloo) |
| Abstract: | Text data, such as answers to open-ended questions, are sometimes ignored because they are hard to analyze. Our Stata command ngram turns text into hundreds of variables using the "bag of words" approach. Broadly speaking, each variable records how often the corresponding word or word sequence occurs in a given text. This is more useful than it sounds. The program supports text in 12 European languages. (Schonlau, M, Guenther, and N Sucholutsky 2017) |
| Date: | 2020–08–20 |
| URL: | http://d.repec.org/n?u=RePEc:boc:scon20:10&r=all |
| By: | Kye Lippold (UC San Diego) |
| Abstract: | I present an applied example of blending theory and data using Stata 16's new Python integration. The SymPy library in Python makes a wide range of symbolic mathematical tools available to Stata programmers. For a recent project, I used theory and SymPy to derive a relationship between two labor supply elasticities in a structural model and separately used Stata to generate reduced-form estimates of these elasticities. I then used the Stata Function Interface to directly plug the empirical Stata estimates into my SymPy model, allowing easy and reproducible estimation of the theoretical relationship of interest. I discuss these methods and provide code for use by other researchers. |
| Date: | 2020–08–20 |
| URL: | http://d.repec.org/n?u=RePEc:boc:scon20:22&r=all |
| By: | Angelo Cozzubo (University of Chicago) |
| Abstract: | In Peru, 55% of the population considers insecurity as the country's main problem. The present study seeks to contribute to the understanding of the social costs of crime in Peru by measuring the impact of patrimonial crime on trust in public institutions, using victimization surveys and censuses of police stations and municipalities and using the newly implemented machine-learning techniques in Stata combined with propensity score matching. Results: reduction of 3 percentage points (pp.) in the probability of trusting in the police and Serenazgo in the short term and 2 pp. in judicial power in the long term. Female victims would lose more confidence in Serenazgo and the Public Ministry. Robustness in the presence of unobservables, different pairings, and falsification tests, which would suggest potential causal character. |
| Date: | 2020–08–20 |
| URL: | http://d.repec.org/n?u=RePEc:boc:scon20:27&r=all |