Automatic neural network modeling for univariate time series

Artificial neural networks (ANNs) are an information processing paradigm inspired by the way the brain processes information. Using neural networks requires the investigator to make decisions concerning the architecture or structure used. ANNs are known to be universal function approximators and are capable of exploiting nonlinear relationships between variables. This method, called Automated ANNs, is an attempt to develop an automatic procedure for selecting the architecture of an artificial neural network for forecasting purposes. It was entered into the M-3 Time Series Competition. Results show that ANNs compete well with the other methods investigated, but may produce poor results if used under certain conditions.     

基于反向传播神经网络的煤矿建筑工程造价估算研究

人工神经网络通过模拟人脑或生物神经网络基本特性,具有较强的自组织和自学习能力,可以从大量的数据中提取内在规律,构造出自变量和因变量二者之间的映射关系,从而实现复杂非线性现象的有效模拟和预测。研究表明人工神经网络方法与其他方法相比在工程造价特征因素非线性和动态性描述方面具有潜在优势。     

Models of representation of social mobility and inequality systems. A neural network approach

This paper reports on the results of the application of an innovative technique, i.e. neural network models, to mobility data. Our primary aim is to show that the technique is more flexible than traditional statistical modeling, and that it entails less strong methodological assumptions concerning the phenomenon which they are intended to represent. Two kinds of networks have been applied: heteroassociative networks, used for prevision and class membership recognition; and autoassociative networks, used for simulation tasks. Results obtained from experiments with neural networks on Italian data are highly consistent with the body of knowledge derived from previous classical analysis. The explicative power of neural network models proved to be higher than that of path analysis given their capacity to uncover any kind or relation between variables, whether linear or nonlinear. When compared to log-linear models, they enable the reconstruction of mobility processes within a global frame, controlling all relevant variables at once.     

Machine learning,artificial neural networks and social research

Quality & Quantity - Machine learning (ML), and particularly algorithms based on artificial neural networks (ANNs), constitute a field of research lying at the intersection of different...     

The concept of individual semantic maps in clinical psychology: a feasibility study on a new paradigm

In this paper we propose a new technology able to map the underlying connection scheme among several psychological variables in a single individual. Nine patients with chronic heart failure underwent at regular intervals, two electronic questionnaires to evaluate depression (STAI—short form) and anxiety (STAY-6). Individual semantic maps were developed by Auto Contractive Map, a new data mining tool based on an artificial neural networks acting on the small data set formed by questionnaires items applied serially along time. The clinical psychologist involved in the clinical evaluation of the cases was asked to score the consistency between the information emerging from the graph depicting the structure of the main relationships among items and the clinical picture resulting from the psychological colloquium. All cases reported overall judgments of a good consistency suggesting that the mathematical architecture of the system is able to capture in the dynamics of items value variations through time the underlying construct of the patient psychological status. This technology is promising in remote monitoring of patients’ psychological condition in different settings with the possibility to implement personalized psychological interventions.     

新企业会计准则下盈余结构的价值相关性——来自沪市A股的经验证据

基于2007年沪市A股中报数据,从相对和增量价值相关性两个角度研究了盈余结构以及具体盈余项目与股票回报之间的关系。研究发现,全面收益的价值相关性水平不如净利润,投资者对利润表的关注程度远高于所有者权益变动表;同时,对总括盈余进行分解能显著增加盈余对股票回报的解释力,除全面收益中部分项目外,各分解项目对股票回报均有显著的增量价值相关性,这表明披露结构性盈余项目有利于投资者判断公司价值。     

On the inclusion of bivariate marked point processes in graphical models

Graphical models use graphs to represent conditional independence relationships among random variables of a multivariate probability distribution. This paper introduces a new kind of chain graph models in which nodes also represent marked point processes. This is relevant to the analysis of event history data, i.e. data consisting of random sequences of events or time durations of states. Survival analysis and duration models are particular cases. This article considers the case of two marked point processes. The idea consists of representing a whole process by a single node and a conditional independence statement by a lack of connection. We refer to the resulting models as graphical duration models.     

Selecting explanatory factors of voting decisions by means of fsQCA and ANN

This investigation applies fuzzy-set qualitative comparative analysis (fsQCA) and an artificial neural networks method (ANN) with the aim of addressing the determinants of votes regarding managerial proposals presented in corporate meetings. The data refer to companies in the United States banking industry and they cover the period from 2003 to 2013. The results show that the variables that contribute to explain the voting support have changed over time. Thus, during the 2003–2006 sub-period the number of funds voting appears as the most clearly outstanding variable. On the contrary, in the 2007–2009 sub-period there is a heterogeneous set of explanatory features that includes the total volume of assets, the leverage ratio and the return on assets ratio, among others, as the most remarkable factors. Finally, in the 2010–2013 sub-period, there are no specific features or combinations that contribute to voting support, indicating that the explanatory factors are yet to be consolidated after the financial downturn.     

Probabilistic energy forecasting using the nearest neighbors quantile filter and quantile regression

Parametric quantile regression is a useful tool for obtaining probabilistic energy forecasts. Nonetheless, traditional quantile regressions may be complicated to obtain using complex data mining techniques (e.g., artificial neural networks), since they are trained using a non-differentiable cost function. This article presents a method that uses a new nearest neighbors quantile filter to obtain quantile regressions independently of the data mining technique utilized and without the non-differentiable cost function. This method is subsequently validated using the dataset from the 2014 Global Energy Forecasting Competition. The results show that the method presented here is able to solve the competition’s task with a similar accuracy to the competition’s winner and in a similar timeframe, but requiring a much less powerful computer. This property may be relevant in an online forecasting service for which the fast computation of probabilistic forecasts using less powerful machines is required.     

上市公司企业利润质量分析——来自国内某大型空调制造企业的实践

本文以利润表为主要分析对象,以定量分析为主要分析方法,对我国空调市场某领导企业(文中简称G公司)进行了利润质量分析与评价,以期发现目前我国上市公司利润表中存在的粉饰公司业绩、扭曲利润质量的手段与途径。     

Best classification algorithms in peer-to-peer lending

A proper credit scoring technique is vital to the long-term success of all kinds of financial institutions, including peer-to-peer (P2P) lending platforms. The main contribution of our paper is the robust ranking of 10 different classification techniques based on a real-world P2P lending data set. Our data set comes from the Lending Club covering the 2009–2013 period, which contains 212,252 records and 23 different variables. Unlike other researchers, we use a data sample which contains the final loan resolution for all loans. We built our research using a 5-fold cross-validation method and 6 different classification performance measurements. Our results show that logistic regression, artificial neural networks, and linear discriminant analysis are the three best algorithms based on the Lending Club data. Conversely, we identify k-nearest neighbors and classification and regression tree as the two worst classification methods.     

基于神经网络的股市资产收益率识别

本文以BP神经网络来识别中国股市不同行业未来的资产收益率,来检验财务数据的有效信号的假定,输入神经元选取了安全性指标、盈利性指标、成长性指标、现金流量结构分析指标四大类指标,共15个最为经典的财务指标,用以预测不同行业的未来超额收益率。预测结果显示,平均预测误差为0.7%,表明神经网络在分析行业未来盈利能力有很强的分类功能。模型分析证实财务数据在行业层面上有十分强烈的信号。同时,预测方法为投资者掌握未来投资资金流向有实用价值。     

基于人工神经网络的非标产品制造成本估算研究

大型和形状复杂的非标产品的生产成本,是很难估算的。目前采用的人工方法估算的误差较大。当成本估算用于招标目的时,这样的误差将产生不利的后果。成本高估承担的风险,是使公司缺乏竞争力,失去客户,而低估了成本,赢得了合同,但招致财务的损失。因此,一个精确的产品生产成本估算方法,对项目管理而言至关重要。本文应用神经网络建立的参数模型,针对大型和形状复杂的非标产品的生产成本估算和目前采用的手动估算相比,精确度有了很大提高。     

全面收益观发展动态及其在我国会计准则体系中的应用

本文认为,全面收益概念的出现标志着人们已从更广阔的意义上关注收益信息的披露。适应全面收益观对财务业绩报告进行改革是国际趋势,我国新发布的新企业会计准则体系也反映出了全面收益观的应用,引入了利得和损失概念,增加了所有者权益变动表,并对利润表的结构和内容进行了改革。     

Identifying Decision Useful Information with the Matrix Format Income Statement

We conduct an experiment to investigate the potential benefits of an alternative format for the income statement, the matrix format, initially developed by the International Accounting Standards Board (IASB) and UK Accounting Standards Board in their joint project on performance reporting. Sophisticated financial statement users (financial analysts and professional accountants) and less sophisticated financial statement users (MBA students) were asked to extract information from a set of financial statements that included an income statement either in the IAS 1 format or in the matrix format. We find that the matrix format improves the accuracy with which users extract financial information. This result is driven by greater accuracy, for all user groups, on “below‐the‐line” items. Furthermore, despite lack of familiarity with the matrix format, its use did not appear to affect the time taken, the ease of extracting financial information, or users' task completion confidence; further experience with the matrix format could lead to benefits along these lines as well. Our findings may assist the FASB and IASB in their joint project on financial statement presentation.     

基于BP神经网络实现的B样条插值在系统动力学模型的应用

在系统动力学建模的过程中,系统变量之间的关系难以确定,传统的建模方法存在着很大的主观性,利用BP神经网络的方法则可以避免这个问题。而在建立BP神经网络过程中需要大量的学习样本,然而通常采集到的数据往往是不足的。用3次B样条函数对历史数据进行插值的方法构建系统中状态变量的学习样本,解决BP神经网络模型中学习样本不足的缺点,能更好的反应变量之间的非线性映射关系。     

Supplier relationship portfolio management: A social exchange perspective

Despite the importance of relationship portfolios, it's unclear how a buying firm's differential investment in its suppliers affects the distribution of its supplier relationships and the supplier-provided benefits that result. Drawing from social exchange theory (SET), we assess the sequential linkages among supply management practices, supplier relationship sets that vary in closeness, and relational benefits. Empirically, we adopt a multi-methodological approach that combines abductive case-based and deductive survey-based research. In our case-based approach, interview responses from 34 professionals within a global Tier 1 automotive manufacturer (MFGR) and four of its suppliers, open-ended survey responses from 56 buyers and 86 engineers within MFGR, documentary evidence, and direct observations facilitate the operationalization of supply management practices and relationship closeness constructs. The survey-based study integrates case-based findings and uses response data from sales managers within 292 suppliers to MFGR and matched supplier performance data from MFGR to test a theoretical model of social exchange. In a multi-step process, we apply cluster analysis, multinomial logistic regression, ANOVA, and multiple regression to this aggregated dataset to (1) identify three distinct sets of supplier relationships that are distributed along a relationship closeness continuum, (2) show how specific supply management practices affect the composition of supplier relationship sets that comprise a buying firm's portfolio, and (3) demonstrate how supplier-provided benefits differ across supplier relationship sets. Our results validate the utility of SET as applied to supplier portfolio management and provide insights into buyers' actions that drive closer relationships, minimize risk, and maximize benefits across a supplier portfolio.     

债务法下所得税会计研究

债务法分为资产负债表债务法和损益表债务法。本文对两种不同的债务法对所得税的会计处理进行了研究,探讨了资产负债表债务法与损益表债务法的特点,并对理解和应用《企业会计准则第18号——所得税》提出了建议。     

Relationships among levels of government support,marketing activities,and financial health of nonprofit performing arts organizations

• This exploratory study researches and analyzes the empirical effects of financial levels of government support and marketing practices on the financial health of nonprofit performing arts organizations. Declining government subsidies and increasing competition from non-traditional sources have presented these organizations with unique opportunities and challenges that call for a market-centered, as well as an art-centered approach. Financial health is critical for the long-term success of a nonprofit performing arts organization. Few empirical studies have examined the interrelationships between these key variables. This study analyzes a sample of 63 American professional symphony orchestras employing 20 years of data. The correlations among financial levels of government support and marketing activities were positive and significant, the correlations among levels of government support and financial health were negative and significant, and the correlations between levels of marketing activity and financial health were negative and varied in significance. Causal analyses were less conclusive, but significant causal relationships were found for large symphony orchestras, indicating that segmentation research may be warranted. Implications and opportunities for future research are presented which have potential application for government agencies, academic researchers, and arts organization managers, boards of directors, and donors.

Copyright © 2006 John Wiley & Sons, Ltd.     

构建会计报表分析方法体系的构想

会计报表分析是了解公司财务状况和经营成果的有力武器。但目前在会计报表分析方法上,没有一个规范的理论体系,会计报表分析的作用很难发挥出来。本文针对以上问题,借鉴统计分析理论,提出了多种会报表的分析方法,并对各种分析方法的含义及其应用范围进行了界定,从而形成一套既具有理论基础,又有很强的实用价值的会计报表分析方法体系。