A comparison of statistical tests for the adequacy of a neural network regression model

An integral part of econometric practice is to test the adequacy of model specifications. If a model is adequately specified, it should not leave interesting features of the data-generating process in the errors. Despite the common tradition, the importance of diagnostic checking as a safeguard against mis-specification has only recently been recognized by neural network (NN) practitioners, possibly because this type of semi-parametric methodology was not originally designed for economic and financial applications. The purpose of this paper is to compare a number of analytical statistical testing procedures suitable to diagnostic checking on a neural network regression model. We present the standard Lagrange multiplier (LM) testing framework designed under the assumption of identically distributed disturbances and also examine two modifications that are robust to heteroskedasticity in errors. One modification also gives the researcher an opportunity to incorporate information concerning the volatility structure of the data-generating process in the testing procedure. By means of a Monte Carlo simulation, we investigate the performance of these tests under GARCH-type heteroskedasticity in errors and various distributional assumptions. The results show that although the primary concern of the researcher may be to design a regression model that accurately captures relations in the mean of the conditional distribution, developing a good approximation of the underlying volatility structure generally increases the efficiency of tests in detecting non-adequacy of a NN model.     

Grey Relational Analysis and Neural Network Forecasting of REIT returns

This study employs the Grey Relational Analysis (GRA) and Artificial Neural Network (ANN) to measure the impact of key elements on the forecasting performance of real estate investment trust (REIT) returns. To manage risks from a real estate price bubble, the findings of GRA suggest that the REIT is best influenced by industrial production index, lending rate, dividend yield, stock index and its own lagged performance. Consequently, this paper adjusts the parameters from GRA and inserts the key elements into the fitted ANN model by comparing the learning effect of the Back-propagation Neural Network (BPN). This study found that the ranking provided by the GRA is significant in correcting prediction errors using the learning outcome of the BPN. The neural network model proved to minimize error function and was able to adjust weighted values in order to enhance prediction accuracy.     

Infrastructure investment and Spanish economic growth, 1850-1935

This paper analyzes the impact of infrastructure investment on Spanish economic growth between 1850 and 1935. Using new infrastructure data and VAR techniques, this paper shows that the growth impact of local-scope infrastructure investment was positive, but returns to investment in large nation-wide networks were not significantly different from zero. Two complementary explanations are suggested for the last result. On the one hand, public intervention and the application of non-efficiency investment criteria were very intense in large network construction. On the other hand, returns to new investment in large networks might have decreased dramatically once the basic links were constructed.     

Principles for the design of management control systems in knowledge networks: Experiences involving the European technology networks

The aim of this paper is to analyse technology-knowledge networks, especially the aspects relating to management control. The management of technology knowledge in a network is characterised by a number of special features that are typical of this form of organization, on the basis of which, we point out the main obstacles and disadvantages that condition the achievement of efficiency of their management. We will analyse management in these networks in light of the experience of various European institutions that actively participate in networks for the development of technological projects. The principal mechanisms used for the control of management will be obtained from the empirical study. Finally, on the basis of the empirical verification, we will consider the development of certain general principles for control in order to achieve greater efficiency in the management of knowledge in these networks.     

技术交易规则预测能力与收益率动态过程——基于Bootstrap方法的实证研究

本文对我国股票市场技术交易规则预测能力进行了实证检验,发现移动平均规则所产生的买入区间收益率更大而波动率却更小,卖出区间的收益率为负而波动率却更大。运用自举(Bootstrap)方法检验发现,四种常用的收益率线性模型均不能解释买卖出区间收益率与波动率所表现出的非对称现象,尤其无法解释卖出区间收益率为负的现象。为此,本文通过人工神经网络方法,将条件异方差结构引入到现有的收益率非线性模型,发现该模型能更好地解释买卖出区间收益率与波动率模式,表明收益率动态过程中存在非线性特征。     

BP神经网络在供应链绩效评价中的应用

首先区分不同主体,权衡成本与收益分层次选择指标,基于流程构建一个供应链绩效评价的指标体系;然后介绍了BP神经网络原理及模型,最后探讨了BP神经网络在供应链绩效评价中的应用。     

基于人工神经网络第三方物流企业的物流中心选址研究

建立了第三方物流企业的物流中心选址评价指标体系，探讨并应用人工神经网络解决第三方物流企业的物流中心选址评价问题，并给出了神经网络的评价模型。实例证明，该方法是可行的。     

基于ANN—BP模型的电子商务信用风险形成思路

企业生存风险的识别的本质是确定企业生存风险识别的模式和影响企业生存风险状态的各指标权重,但是众多影响因素间不存在确定的函数关系表达式,并且各指标权重的确定也相当复杂。人工神经网络（ANN）基于并行处理机制从结构上对人类的思维过程进行模拟,从而能实现人类思维的某些功能。人工神经网络可以实现任意形式的映射,这就为企业生存风险识别提供了一种新的思路。基于人工神经网络（ANN）的电子商务信用风险模式识别,能够充分利用样本电子商务信用风险的有关信息,通过高度的非线性映射,揭示感知信用风险与其相关影响因素即主要诱因的内在作用机理,从而从根本上克服了感知信用风险测度或识别中建模及其求解的困难。     

Structural analysis and forecast of gold price returns

Gold has multiple attributes and its price is affected by various factors in the market. This paper studies the dynamic relationship between the gold price returns and its affecting factors. Then we use the STL-ETS, neural network and Bayesian structural time series model to predict the gold price returns, and compare their performance with the benchmark models. The results show that the shocks of crude oil returns and VIX have the positive effect on gold price returns, the shocks of the US dollar index have the negative effect on gold price returns. And the fluctuation of gold price returns mainly depends on crude oil price returns shocks. STL-ETS model can accurately fit the fluctuation trend of the gold price returns and improve prediction accuracy.