Abstract:Considering a supply chain consisting of a manufacturer，a retailer and a third-party B2B online spot market，we research the manufacturer’s optimal bidding strategy for options and the retailer’s optimal procurement strategy in a Stackelberg game with the manufacturer as the leader．In our model，we measure the liquidity of B2B online spot market by a parameter which means the probability of transaction success．We dis_x005fcuss two situations: manufacturer with adequate capacity and the manufacturer with limited capacity，and expand our conclusions to the special case of uniform distribution．Finally，we analyze the impacts of spot market liquidity on behaviors and profits of all participants in the supply chain through the numerical examples．

Abstract:The issue of the asking price ( AP) before trading has received much attention from economists．Most of the existing literature argues that AP can be used as a signal of the seller’s reservation price or pro_x005fduction cost，which is the private value of the seller．Little attention is paid to the relationship between the AP and the quality of an asset，which is a common value of the seller．We examine the signaling mechanism of the AP under bilateral asymmetric information．We find that，given that the trade opportunity shrinks with increas_x005fing quality，there are two contrasting effects with the change of the AP: the price effect and the trade effect．The marginal rate of substitution between the trade effect and the price effect is higher for the seller with a lower quality asset．This means that the single-crossing condition is satisfied in this bilateral asymmetric information game．Given the boundary condition is satisfied，there is a unique separating equilibrium．A seller with a lower quality has an incentive to signal the quality with a lower AP，and it is reasonable to expect a higher quality from assets with a higher AP．All possible trades will be accomplished in this unique separating equilibrium despite the informational asymmetry．

Abstract:The effectiveness of revenue management methods is based on the assumption that consumers will make purchase decisions in accordance with the behavior rules the sellers have presupposed．Current revenue management models either consider customers to be myopic or to be rational．In this paper，these two views to_x005fwards the consumer behavior are questioned．This paper studies the behavior characteristics of airline passengers within the prospect theory framework using Chinese empirical data．The empirical study shows that the passengers’purchasing behavior has bounded-rationality characteristics called LA-DS-RE-RD，ie，loss aversion，diminishing sensitivity，reflection effect and reference dependence．Furthermore，this paper uses KANO model to select the key elements that affect the passengers’utility judgments and constructs a purchasing utility measurement framework．The study shows that，in revenue management situations，the utility judgement process before making purchasing decisions not only refers to the traditional expected utility system but is adjusted by reference utility．

Abstract:This paper focuses on how to conduct effective and real-time rescheduling at an open shop subject to random Machine UnAvailability for a Duration ( MUAD) and its concurrent Compressions of Processing Times ( CPT) ． In this study，efficiency is measured by the makespan，while the stability measure is associat_x005fed with the sequence deviation and ending time deviation．With three typical rescheduling strategies in the lit_x0002_erature ( i.e.，right-shift rescheduling，affected operations rescheduling and total rescheduling) ，three specific approaches named sRSR，sAOR and sTR_GOS are proposed and implemented for the concerned MUADCPT disruption．In extensive experiments，by randomly generating various rescheduling scenarios，three initializations of previous schedules are examined and three rescheduling approaches are independently tested．The results statistically reveal that: 1) sAOR is highly recommended when the MUAD disruption lasts for a short duration and Interrupt-Resume mode is adopted; 2) sTR_GOS is relatively desirable in case of late MUAD disruption，low compression rate，small instance size and Interrupt-Repeat mode．

Abstract:Repetitive scheduling method (RSM) and the network models have their own advantages in project scheduling．This paper proposes a method for converting the RSM to the network model and compares the two methods in terms of critical paths，activity criticality and floats．The proposed method establishes a rule for converting all kinds of activities and constraints in RSM to the network model．It can represent the resource continuity which could avoid the difference between RSM and the converted network model in total durations，critical paths and floats in literatures．Besides，it can also display the distance constraint and progress in space and resolves the problem that the network model could not show distance constraints and space conflicts．A re_x005fpetitive project is used as an example of conversion and comparison．The analysis of activity criticality reveals the reason why existing methods could not completely convert the RSM to network models．This method can provide the project managers with a convenient tool to transfer the RSM into the equivalent network model．

Abstract:Based on the theory of bank capitalization and risk-taking ，this paper tests the cyclicality hypothe_x005fsis and examines relationship between banking capital and risk endogenous adjusting behaviors under the capital adequacy requirement by constructing simultaneous equations model with economic cycle，market structure as macro environment variables and loan provision，expenses management as bank specific variables． System Generalized Method of Moments is employed to estimate the model with annual data of the 15 major commercial banks in China over the period between 2004 and 2010．The results show that the association between the change in capital and risk is negative; the bank capitalization and risk-taking behaviors are countercyclical; the reduction of bank concentration has significant positive impacts on capital increases and risk reduction; the capital adequacy is effective in increasing the capital adequacy ratio，but not in reducing the portfolio risk．

Abstract:Considering the facts that the data are not high frequent ones and the business lines are more than two，the copula models used to construct the joint distributions in market risk measurement are not completely applied to the underwriting risks measurement，this paper constructs a comonotonicity model to tentatively dis_x005fcuss how to measure the economic capital for underwriting risks combined with the characteristics of the underwriting risks．The paper uses a real property insurance to show the process of the assessment process of the underwriting risks．In the empirical analysis，the paper also compares the comonotonicity model and the copula model．Empirical results show that both the comonotonicity model and the copula model can measure the diversification of the business lines． However，the comonotonicity model can get more accurate results than the copula model．

Abstract:The article mainly researches the setting of the margin levels of Shanghai and Shenzhen 300 stock index futures and the determination of the default rate．Firstly，it applies the methods of both Hill estimation and VaR-x estimation to solve the estimated values of the tailing exponential of the full-samples，the left-tail and the right-tail，and finds the margin levels are 3.571 7% and 5.334% respectively．Secondly，it makes a backtracking test by comparing the estimated values of the margin levels with the actual historic pricing volatil_x005fity and discovers that the margin levels derived from neither Hill estimation nor VaR-x estimation can cover 99% of the asset pricing volatilities under the assumption that the default rate is 1% .Then it considers the default rate being 2% ，3% and 4% respectively．The results show: ( 1) .when the default rate equals 3% ，the margin level derived from Hill estimation can cover more than 97% of the pricing volatilities; when the default rate is within 1% － 3% ，the margin level derived from VaR-x estimation is reasonable．There is no significant difference between the left-tail and the right-tail of the tailing exponential got by the Hill estimation．Therefore，it does not need to set different margin levels for short positions and the long positions． But accord_x005fing to VaR-x estimation，the margin level of left-tail is significantly lower that the margin level of right-tail，so different margin levels are needed for short positions and the long positions．