PAMR: Passive aggressive mean reversion strategy for portfolio selection

This article proposes a novel online portfolio selection strategy named “Passive Aggressive Mean Reversion” (PAMR). Unlike traditional trend following approaches, the proposed approach relies upon the mean reversion relation of financial markets. Equipped with online passive aggressive learning technique from machine learning, the proposed portfolio selection strategy can effectively exploit the mean reversion property of markets. By analyzing PAMR’s update scheme, we find that it nicely trades off between portfolio return and volatility risk and reflects the mean reversion trading principle. We also present several variants of PAMR algorithm, including a mixture algorithm which mixes PAMR and other strategies. We conduct extensive numerical experiments to evaluate the empirical performance of the proposed algorithms on various real datasets. The encouraging results show that in most cases the proposed PAMR strategy outperforms all benchmarks and almost all state-of-the-art portfolio selection strategies under various performance metrics. In addition to its superior performance, the proposed PAMR runs extremely fast and thus is very suitable for real-life online trading applications. The experimental testbed including source codes and data sets is available at .     

The ability of directed tests to predict software quality

Many statistical methods for estimating software quality rely on representative testing: they assume a program is tested in an environment that simulates the environment where it will be used. Often, however, a software tester’s aim is to uncover defects as soon as possible, and representative testing may not be the best way to do this. Instead, tests are often selected according to some plan that is believed to result in an efficient but thorough examination of the software’s behavior. This raises the question of how practical measurements of software quality, like software probability‐of‐failure, can be obtained from directed testing. In this paper, we discuss some factors affecting the ability of directed tests to predict software quality when quality is measured in the environment where the software operates, but the directed tests do not simulate that environment. We consider a number of ways to measure the power of a directed test method, and show how these affect the tester’s ability to predict software quality.     

Compositional Analysis of C/C++ Programs with VeriSoft

This paper describes how the state space exploration ool VeriSoft can be used to analyze parallel C/C++ programs compositionally. VeriSoft is employed for two analyses: transition traceanalysis and assume/guarantee reasoning. Both analyses are compositional in the sense that the behaviour of a parallel program is determined in terms of the behaviour of its constituent processes. While both analyses have traditionally been carried out with “pencil and paper”, the paper demonstrates how VeriSoft can be used to automate them. In the context of transition trace analysis, the question whether a given program can exhibit a given trace is addressed with VeriSoft. To implement assume/guarantee reasoning, VeriSoft is used to determine whether a given program satisfies a given assume/guarantee specification. Since VeriSoft’s state space exploration is bounded and thus not complete in general, our proposed analyses are only meant to complement standard reasoning about parallel programs using traces or assume/guarantee specifications. For instance, a successful analysis does not always imply the general correctness of an assume/guarantee specification. However, it increases the confidence in the verification effort. On the other hand, an unsuccessful analysis always produces a counterexample which can be used to correct the specification or the program. VeriSoft’s optimization and visualization techniques make the analyses relatively efficient and effective.     

Reaching pareto-optimality in prisoner’s dilemma using conditional joint action learning

We consider the learning problem faced by two self-interested agents repeatedly playing a general-sum stage game. We assume that the players can observe each other’s actions but not the payoffs received by the other player. The concept of Nash Equilibrium in repeated games provides an individually rational solution for playing such games and can be achieved by playing the Nash Equilibrium strategy for the single-shot game in every iteration. Such a strategy, however can sometimes lead to a Pareto-Dominated outcome for games like Prisoner’s Dilemma. So we prefer learning strategies that converge to a Pareto-Optimal outcome that also produces a Nash Equilibrium payoff for repeated two-player, n-action general-sum games. The Folk Theorem enable us to identify such outcomes. In this paper, we introduce the Conditional Joint Action Learner (CJAL) which learns the conditional probability of an action taken by the opponent given its own actions and uses it to decide its next course of action. We empirically show that under self-play and if the payoff structure of the Prisoner’s Dilemma game satisfies certain conditions, a CJAL learner, using a random exploration strategy followed by a completely greedy exploitation technique, will learn to converge to a Pareto-Optimal solution. We also show that such learning will generate Pareto-Optimal payoffs in a large majority of other two-player general sum games. We compare the performance of CJAL with that of existing algorithms such as WOLF-PHC and JAL on all structurally distinct two-player conflict games with ordinal payoffs.     

Multicriterial investment problem in conditions of uncertainty and risk

Lower and upper attainable bounds for stability radius of efficient solution to multicriterial Boolean portfolio optimization problem with Savage’s minimax risk criterion are obtained.     

Safety-first portfolio selection

A. D. Roy’s safety-first (SF) approach to financial portfolio optimization is improved. Safety first means the minimization of the probability of negative returns. The improvement concerns a better estimation of the negative return probabilities by means of mean excess return risk functions. The search for the optimal SF-portfolio is similar to Roy’s geometric method but the efficient frontier is different. In case of a finite number of scenarios, the SF-portfolio selection problem is reduced to a mixed linear Boolean programming problem.     

Amdahl’s Law Revisited for Single Chip Systems

Amdahl’s Law is based upon two assumptions – that of boundlessness and homogeneity – and so it can fail when applied to single chip heterogeneous multiprocessor designs, and even microarchitecture. We show that a performance increase in one part of the system can negatively impact the overall performance of the system, in direct contradiction to the way Amdahl’s Law is instructed. Fundamental assumptions that are consistent with Amdahl’s Law are a heavily ingrained part of our computing design culture, for research as well as design. This paper points in a new direction. We motivate that emphasis should be made on holistic, system level views instead of divide and conquer approaches. This, in turn, has relevance to the potential impacts of custom processors, system-level scheduling strategies and the way systems are partitioned. We realize that Amdahl’s Law is one of the few, fundamental laws of computing. However, its very power is in its simplicity, and if that simplicity is carried over to future systems, we believe that it will impede the potential of future computing systems.     

Intelligent forecasting models-selection system for the portfolio internal structure change

In this study, an intelligent forecasting models-selection system for refining portfolio structural estimation is proposed selecting different forecasts time series models, as well as the contents’ trend with refining the risk-return matrices of components. Based on the four inference rules in intelligent selection mechanism, the support system seeks to find the appropriate model solutions satisfying the tracking for the behavior of indices prices in portfolio optimization. The feasibility of the system is verified with a practical simulation experiment. The experimental results show that, for all examined investment assets, the presented system is an efficient way of solving the portfolio internal structure change problem. In addition, we also find that the presented system can also be used as an alternative method for evaluating various forecasting models. By means of global major market as the empirical evidences of portfolio contents, it will show that the proposed system can serve as improving efficient frontier of a portfolio.     

Functional forms in studies of reputation in online auctions

Many studies have examined how various factors affect prices in online auctions. These studies assume that the relationship between price and the seller’s reputation take a variety of functional forms, most frequently linear or linear-log. Others divide the sellers into categories by their reputations, and control for dummy variables indicating the seller’s category. Identifying the correct functional form is a critical issue for research on any topic involving online auctions. Studies that assume the wrong functional form run the risk of generating biased and inconsistent estimates of the effect of their variables of interest. In this study, the price-reputation relationship is estimated under each of these functional forms using data from auctions of three different products. The estimated effect of reputation on price is substantially larger when using a categorical specification. The models are then subjected to specification tests which suggest that the categorical model is the most appropriate choice.     

Local and Global Dynamics in an Overlapping Generations Model with Endogenous Time Discounting

The aim of this paper is to develop an overlapping generations model with endogenous time preference. We assume that the individual’s rate of time preference is determined by the labour supply at the economy wide level. We show that when the rate of impatience is constant, a unique steady state exists, which is saddle-point stable for several constellations of parameter values. In contrast, when the rate of impatience is determined by the aggregate labour supply, multiple steady may exist, and the equilibrium dynamics may exhibit (local or global) indeterminacy and complex behaviour.     

Modeling complex longitudinal consumer behavior with Dynamic Bayesian networks: an Acquisition Pattern Analysis application

Longitudinal consumer behavior has been modeled by sequence analysis. A popular application involves Acquisition Pattern Analysis exploiting typical acquisition patterns to predict a customer’s next purchase. Typically, the acquisition process is represented by an extensional, unidimensional sequence taking values from a symbolic alphabet. Given complex product structures, the extensional state representation rapidly evokes the state-space explosion problem. Consequently, most authors simplify the decision problem to the prediction of acquisitions for selected products or within product categories. This paper advocates the use of intensional state definitions representing the state by a set of variables thereby exploiting structure and allowing to model complex, possibly coupled sequential phenomena. The advantages of this intensional state space representation are demonstrated on a financial-services cross-sell application. A Dynamic Bayesian Network (DBN) models longitudinal customer behavior as represented by acquisition, product ownership and covariate variables. The DBN provides insight in the longitudinal interaction between a household’s portfolio maintenance behavior and acquisition behavior. Moreover, it exhibits adequate predictive performance to support the financial-services provider’s cross-sell strategy comparable to decision trees but superior to MulltiLayer Perceptron neural networks.     

On-line service scheduling

This paper is concerned with a scheduling problem that occurs in service systems where customers are classified as ‘ordinary’ and ‘special’. Ordinary customers can be served on any service facility, while special customers can be served only on the flexible service facilities. Customers arrive dynamically over time and their needs become known upon arrival. We assume any service, once started, will be carried out to its completion. In this paper, we study the worst-case performance of service policies used in practice. In particular, we evaluate three classes of service policies: policies with priority, policies without priority, and their combinations. We obtain tight worst-case performance bounds for all service policies considered.     

Exploring the feasibility of web form adaptation to users’ cultural dimension scores

With many daily tasks now performed on the Internet, productivity and efficiency in working with web pages have become transversal necessities for all users. Many of these tasks involve the inputting of user information, obligating the user to interact with a webform. Research has demonstrated that productivity depends largely on users’ personal characteristics, implying that it will vary from user to user. The webform development process must therefore, include modeling of its intended users to ensure the interface design is appropriate. Taking all potential users into account is difficult, however, primarily because their identity is unknown, and some may be effectively excluded by the final design. Such discrimination can be avoided by incorporating rules that allow webforms to adapt automatically to the individual user’s characteristics, the principal one being the person’s culture. In this paper we report two studies that validate this option. We begin by determining the relationships between a user’s cultural dimension scores and their behavior when faced with a webform. We then validate the notion that rules based on these relationships can be established for the automatic adaptation of a webform in order to reduce the time taken to complete it. We conclude that the automatic webform adaptation to the cultural dimensions of users improves their performance.     

绿色供应链的投贷联动融资决策均衡

以一个零售商和一个资金约束的制造商组成的绿色供应链为研究对象,构建了制造商资金约束情况下银行借贷融资、股权融资和组合融资时的融资模型,探讨制造商的资金水平和消费者的绿色偏好对定价、绿色投入水平、产品的市场需求和融资方式选择策略的影响.最后,通过数值分析进行验证.研究发现:消费者绿色偏好与零售价格、批发价格和绿色投入努力...     

General default logic

In this paper, we propose a new nonmonotonic logic called general default logic. On the one hand, it generalizes Reiter’s default logic by adding to it rule-like operators used in logic programming. On the other hand, it extends logic programming by allowing arbitrary propositional formulas. We show that with this new logic, one can formalize naturally rule constraints, generalized closed world assumptions, and conditional defaults. We show that under a notion of strong equivalence, sentences of this new logic can be converted to a normal form. We also investigate the computational complexity of various reasoning tasks in the logic, and relate it to some other nonmonotonic formalisms such as Lin and Shoham’s logic of GK and Moore’s autoepistemic logic.     

Cognitive and synthetic behavior of avatars in intelligent virtual environments

In intelligent virtual environments (IVEs), it is a challenging research issue to provide the intelligent virtual actors (or avatars) with the ability of visual perception and rapid response to virtual world events. Modeling an avatar’s cognitive and synthetic behavior appropriately is of paramount important in IVEs. We propose a new cognitive and behavior modeling methodology that integrates two previously developed complementary approaches. We present expression cloning, walking synthetic behavior modeling, and an autonomous agent cognitive model for driving an avatar’s behavior. Facial expressions are generated using our own-developed rule-based state transition system. Facial expressions are further personalized for individuals by expression cloning. An avatar’s walking behavior is modeled using a skeleton model that is implemented by seven-motion sequences and finite state machines (FSMs). We discuss experimental results demonstrating the benefits of our approach.     

On the bottom summation

We consider summation of consecutive values (φ(v), φ(v + 1), ..., φ(w) of a meromorphic function φ(z), where v, w ∈ ℤ. We assume that φ(z) satisfies a linear difference equation L(y) = 0 with polynomial coefficients, and that a summing operator for L exists (such an operator can be found—if it exists—by the Accurate Summation algorithm, or, alternatively, by Gosper’s algorithm when ordL = 1). The notion of bottom summation which covers the case where φ(z) has poles in ℤ is introduced. The text was submitted by the authors in English.     

Life and Mind

It’s sometimes said, and even more often assumed, that life is necessary for mind. If so, and if A-Life promises to throw light on the nature of life as such, then A-Life is in principle highly relevant to the philosophy of mind and cognitive science. However, very few philosophers have attempted to argue for the relation between life and mind. It’s usually taken for granted. Even those (mostly in the Continental tradition, including some with a following in A-Life) who have insisted on the linkage have stated it rather than justified it. If an evolutionary account of intentionality is acceptable, then perhaps biological life ‘makes room’ for mind. But that claim is problematic, since it’s not clear that the type of self-organization involved in life-as-such must necessarily include evolution. Even if it does, it’s a further step to show that life is strictly necessary for mind.     

Improved bound for stochastic formal correctness of numerical algorithms

We provide bounds on the probability that accumulated errors were never above a given threshold on numerical algorithms. Such algorithms are used, for example, in aircraft and nuclear power plants. This report contains simple formulas based on Lévy’s, Markov’s and Hoeffding’s inequalities and it presents a formal theory of random variables with a special focus on producing concrete results. We select three very common applications that cover the common practices of systems that evolve for a long time. We compute the number of bits that remain continuously significant in the first two applications with a probability of failure around one out of a billion, where worst case analysis considers that no significant bit remains. We are using PVS as such formal tools force explicit statement of all hypotheses and prevent incorrect uses of theorems.