基于注意力机制的LSTM股价趋势预测研究

针对中国股票市场,提出了一种基于注意力机制的LSTM股价趋势预测模型。选取42只中国上证50从2009年到2017年的股票数据为实验对象,根据股票市场普遍认可的经验规则,分别对每个技术指标进行量化处理得到股票涨跌的趋势数据,并和交易数据混合作为预测模型的输入,然后使用基于注意力机制的LSTM模型提取股价趋势特征进行预测。实验结果表明:引入股票离散型趋势数据到预测模型中,能够在已有交易数据和技术指标的基础上提升预测精确度,与传统的机器学习模型SVM和单一的LSTM模型相比,基于注意力机制的LSTM模型具有更好的预测能力。     

Learning to bid in sequential Dutch auctions

We propose an agent-based computational model to investigate sequential Dutch auctions with particular emphasis on markets for perishable goods and we take as an example wholesale fish markets. Buyers in these markets sell the fish they purchase on a retail market. The paper provides an original model of boundedly rational behavior for wholesale buyers׳ behavior incorporating learning to improve profits, conjectures as to the bids that will be made and fictitious learning. We analyze the dynamics of the aggregate price under different market conditions in order to explain the emergence of market price patterns such as the well-known declining price paradox and the empirically observed fact that the very last transactions in the day may be at a higher price. The proposed behavioral model provides alternative explanations for market price dynamics to those which depend on standard hypotheses such as diminishing marginal profits. Furthermore, agents learn the option value of having the possibility of bidding in later rounds. When confronted with random buyers, such as occasional participants or new entrants, they learn to bid in the optimal way without being conscious of the strategies of the other buyers. When faced with other buyers who are also learning their behavior still displays some of the characteristics learned in the simpler case even though the problem is not analytically tractable.     

对项目标后经营的探讨

对于竞争越来越残酷的建筑市场来说,随着工程中标单价越来越低,如何做好标后经营工作就显得尤其重要,这也是施工单位提高经济效益的重要途径。文章就如何做好项目标后经营谈了几点认识。     

The house price determination process: Rational expectations with a spatial context

Applying the rational expectations hypothesis, this essay models the current value of a house as the conditional expectation of the discounted stream of housing services accruing to the owner of the house. The value of housing services is determined by neighborhood effects as well as the physical attributes of the property itself. In the existing hedonic literature, future transactions have not been utilized to describe neighborhood effects. The rational expectations asset pricing model in this study accounts for expected future neighborhood effects as well as observed current neighborhood effects. The reduced form of the rational expectations model is a spatial autoregressive (SAR) model with two spatial lags. After employing the generalized method of moments (GMM) in estimating the spatial asset pricing model, I find that both expected future transactions and prior transactions in the neighborhood are significant. The inclusion of expected future transaction prices in the neighborhood takes into account the influence of expected changes in the community and factors these potential changes into the current house price. This is consistent with forward-looking households. The forward-looking model generates superior out-of-sample prediction performance relative to both the conventional hedonic model without considering neighborhood effects or the standard spatial hedonic model including only past transactions.     

Trading patterns in the European carbon market: The role of trading intensity and OTC transactions

This paper examines the effect of trading intensity and OTC transactions on expected market conditions in the early development period of the European Carbon futures market. Past duration and trading intensity are used as information related order flow variables in modelling time between transactions in two new specifications of Autocorrelation Conditional Duration (ACD) models. This allows for specific investigation of non-linear asymmetric effects on expected duration and the impact of OTC transactions. Evidence is presented of two main types of trading episodes of increased and decreased trading intensity. Both have a significant impact on price volatility, which increases further if an OTC transaction intrudes. OTC transactions also play a dual role. They slow down trading activity in the short term (over the next five transactions) but increase it substantially in the long term (over ten transactions). Both the liquidity and information price impact components increase following an OTC trade, but the information impact is greater. Price volatility calms down faster than liquidity effects following an OTC trade, and this is more pronounced in ECX and in Phase II. The combined evidence points towards increased market depth, efficiency and maturity of the trading environment.     

供应链成员企业的交易成本及其博弈分析

本文分析了供应链成员企业之间交易成本的影响因素,对成员企业之间单次交易的交易成本以及多次交易成本的产生过程进行了深入的剖析。提出时点交易成本、总交易成本、边际交易成本等新概念,给出了描述这些概念的函数曲线图,对多次交易过程进行了博弈分析,进一步导出交易成本的计算式。最后运用文中结论对交易成本进行了实证分析。     

上海市土地使用权出让机制改革研究——基于拍卖理论和上海市土地市场的实证检验

2007年上海市规定住宅用地挂牌出让的现场竞价规则由公开叫价改为一次性书面报价,旨在降低上海市住宅用地的价格.本文在价值相容和可显示信息的假设下,分别建立两阶段升价拍卖模型和混合拍卖模型,刻画改革前后的挂牌机制,指出:竞标者的均衡策略可解释土地挂牌出让中开发商集中在现场竞价阶段内出价,且更倾向在混合拍卖中隐藏信息的行为;当竞价过程中更易于显示信息时,混合拍卖的期望收入低于两阶段升价拍卖,这表明混合拍卖方式能够降低土地价格.上海市的土地交易数据也证明了这一结论.     

Incentivizing resilience in financial networks

When banks extend loans to each other, they generate a negative externality in the form of systemic risk. They create a network of interbank exposures by which they expose other banks to potential insolvency cascades. In this paper, we show how a regulator can use information about the financial network to devise a transaction-specific tax based on a network centrality measure that captures systemic importance. Since different transactions have different impact on creating systemic risk, they are taxed differently. We call this tax a systemic risk tax (SRT). We use an equilibrium concept inspired by the matching markets literature to show analytically that this SRT induces a unique equilibrium matching of lenders and borrowers that is systemic-risk efficient, i.e. it minimizes systemic risk given a certain transaction volume. On the other hand, we show that without this SRT multiple equilibrium matchings exist, which are generally inefficient. This allows the regulator to effectively stimulate a ‘rewiring’ of the equilibrium interbank network so as to make it more resilient to insolvency cascades, without sacrificing transaction volume. Moreover, we show that a standard financial transaction tax (e.g. a Tobin-like tax) has no impact on reshaping the equilibrium financial network because it taxes all transactions indiscriminately. A Tobin-like tax is indeed shown to have a limited effect on reducing systemic risk while it decreases transaction volume.     

The demand for computer services: A disaggregate decision model

Pricing schedules for computer resources have traditionally been based on ‘cost-recovery’ principles. While economists have begun to address pricing based on marginal congestion costs, most models take demand to be exogenous and given. Discrete alternatives are inadequately treated, and aggregation of data precludes any assessment of the impact of transaction size on consumers' decisions. Using disaggregated data, this paper derives empirical results confirming that consumers are strongly influenced by transaction sizes. Simulation experiments demonstrate that price incentives designed to modify the use of computer resources are considerably more effective if the distribution of demand is weighted towards large transactions.     

Past price ‘memory’ in the housing market: testing the performance of different spatio-temporal specifications

Recent methodological developments provide a way to incorporate the temporal dimension when accounting for spatial effects in hedonic pricing. Weight matrices should decompose the spatial effects into two distinct components: bidirectional contemporaneous spatial connections; and unidirectional spatio-temporal effects from past transactions. Our iterative estimation approach explicitly analyses the role of time in price determination. The results show that both spatio-temporal components should be included in model specification; past transaction information stops contributing to price determination after eight months; and limited temporal friction is exhibited within this period. These findings highlight the decidedly non-linear temporal patterns of such information effects.     

Forecasting solar stock prices using tree-based machine learning classification: How important are silver prices?

Solar energy is one of the fastest growing sources of electricity generation. Forecasting solar stock prices is important for investors and venture capitalists interested in the renewable energy sector. This paper uses tree-based machine learning methods to forecast the direction of solar stock prices. The feature set used in prediction includes a selection of well-known technical indicators, silver prices, silver price volatility, and oil price volatility. The solar stock price direction prediction accuracy of random forests, bagging, support vector machines, and extremely randomized trees is much higher than that of logit. For a forecast horizon of between 8 and 20 days, random forests, bagging, support vector machines, and extremely randomized trees achieve a prediction accuracy greater than 85%. Although not as prominent as technical indicators like MA200, WAD, and MA20, oil price volatility and silver price volatility are also important predictors. An investment portfolio trading strategy based on trading signals generated from the extremely randomized trees stock price direction prediction outperforms a simple buy and hold strategy. These results demonstrate the accuracy of using tree-based machine learning methods to forecast the direction of solar stock prices and adds to the broader literature on using machine learning techniques to forecast stock prices.     

我国现行工程招投标体系的制度分析

文章基于我国现行的工程招标投标制度,对现行工程体系中的如串标、假资质、工程多层肢解发包等不合理现象运用制度经济学的原理进行分析,认为现行体系可以从竞价、交易成本、委托代理和契约4方面进行分析,并提出相应对策。     

The influence of seller learning and time constraints on sequential bargaining in an artificial perishable goods market

This paper investigates the formation of prices in a perishable goods market where agents bargain repeatedly through pair-wise interactions. After extensive field observations, we chose to focus on two aspects that seem important to actors of this market: the passage of time and update in judgement when gathering information. The main feature of the market is that a seller bargaining with a buyer has incomplete information about buyer's willingness to pay and is not sure how her trading partner will evaluate an offer or compare it with other options. On the other hand, buyers have limited time to look for goods and cannot meet all possible sellers before making a decision. Hence agents cannot calculate the best price to offer but receive information through limited interactions, and use this information to choose their actions.An agent-based model was built to represent a framework that mimics the observed market institution and where agent's possible behaviors and learning was made as consistent as possible with gathered data. Simulations were run, first for sensitivity analysis concerning main parameters, then to test the dependance of agents’ learning to (a) the time buyers can spend on the market and (b) the frequency of update in learning by sellers. To validate the model, features produced by the simulated market are compared to the stylized facts gathered for negotiation about four goods. We reproduce the main features of the data on the dynamics of offers, transaction prices and agents’ behavior during the bargaining phases.     

基于支持向量机的鸡蛋供应链中价格预警研究

鸡蛋价格的大幅波动对鸡蛋供应链中生产者、经营者和消费者产生严重的影v向,因此鸡蛋供应链中的价格预警问题亟待研究。支持向量机是一种建立在统计学习理论基础上的机器学习方法,能够较好地解决小样本、非线性和局部极小点等问题。文中基于支持向量机方法建立鸡蛋价格预警模型,运用libsvm软件对样本集进行参数寻优、训练和测试。结果表明模型预警结果与实际情况相符,模型可以帮助鸡蛋供应链中的各主体在采购、生产、仓储、销售等方面更好地决策。     

Can money matter for interest rate policy?

In this paper it is shown that money can matter for macroeconomic stability under interest rate policy when transactions frictions are non-negligible. We develop a sticky price model with a shopping time function, which induces the marginal utility of consumption to depend on the (predetermined) stock of money held at the beginning of the period. Equilibrium stability and uniqueness are then ensured by a passive interest rate policy, whereas activeness is associated with an explosive equilibrium. By reacting to changes in beginning-of-period real balances, the central bank can restore stability. Interest rates further depend on lagged real balances even if the central bank acts in an entirely forward-looking way, as under discretionary optimization. If the model is revised such that end-of-period money provides transaction services, money can in principle be neglected for a stabilizing interest rate policy. Discretionary monetary policy is, however, likely to be associated with equilibrium indeterminacy, which can be avoided if interest rates are set contingent on beginning-of-period real balances.     

合理低标价法的实践与探讨

如何认识“合理低标价法”一直是招投标领域讨论的热门话题。基于作者对市场经济的理解以及对招标投标的认识,分析微观市场环境以及宏观经济环境,对未来市场发展趋势的作出阐述,目的是使招投标当事人以及招投标监督管理部门对于“合理低标价法”有一个新的认识,去除对“合理低标价”评标办法的表面的、形式化的争论,把讨论的注意力转移到如何在工作实践中更有效的运用“合理低标价法”上,达到促进市场完善,鼓励竞争,促进整个工程招投标发展的目的。     

谈信息化工程招标投标现状

招标投标是市场经济体制下广泛采用的完成工程交易的主要方式,通过公开招标和邀请招标等不同的招标方式,选择报价低、技术强、信誉好的承包商,有效遏制不正当竞争行为,有利于创造公平竞争的市场环境。当前,我国工程招标投标市场中还存在许多问题,不正当的竞争导致各种豆腐渣工程频现。如何保障双方利益,携手合作、实现共赢是当前急需解决的问题。     

建设工程招投标经济性分析

对工程招投标阶段招标方和投标方的费用构成进行阐述,分析招投标阶段的经济性影响因素,最后提出降低费用经济性的提议。     

Modelling the UK gilt-edged market

In this paper we examine the sectoral demand for UK gilt-edged securities. The Tobin–Markowitz model of portfolio choice generates the prediction that asset holdings should be negatively correlated with the own price, and positively correlated with the prices of major substitutes. In contrast, for all the five major groups of UK gilt holders, we find evidence consistent with a reverse correlation between the own price and market holdings which we argue arises due to the passive revaluation of existing holdings. We examine the empirical evidence using both cointegration analysis of stock holdings and a dynamic model of net transactions.     

The M/G/∞ Queueing Model for Optimistic Concurrency Control

The M / G /∞ queue with Optimistic Concurrency Control (OCC) is a model for a special form of parallel transaction processing in a real-time database. Transactions arrive according to a Poisson process and require some generally distributed execution time. One of the differences with ordinary multi- and infinite-server queueing models is that under OCC the successful completion of one transaction may immediately cause the failure of one or more of the other transactions. This happens if the completing transaction has overwritten a data-item that is in use by another transaction in progress. As soon as this failure is detected the failed transaction is restarted. So the total service time of a transaction consists of its final successful run and the time spent on unsuccessful runs.
In this study we develop an approximation for the distribution of the total service time, and test the approximation against simulation. Although in practice the number of servers is never unlimited, this study provides valuable insight in the asymptotics with respect to the number of servers. The approximation clearly demonstrates the very limited performance gain from an increase of the number of servers.