The feature of group buying is that the price will go down as the accumulated orders are increasing. However, consumers will not know the final price until the end. As a result, consumers can only make decision based on the final price forecast of group buying. The final price forecast might be different depending on different pricing strategies of group-buying models and if there are competitions from posted-price stores. The purpose of this research is to understand how consumers¡¦ internal reference price and final price prediction of group buying would be influenced when facing different price curves in different market competition environment. The difference between the internal reference price and final price prediction of group-buying indicates the consumers¡¦ transaction utility. In addition, if consumers¡¦ perceived transaction utilities affect their intentions and behavior of joining group buying is another research purpose. In this research, there are three different pricing strategies, decreasing, neutral, and increasing based on the initial price, discount size and final price. For the market competition environment, it manipulated by if there are other posted-price stores to be chosen or not. The research result indicates that increasing price curve, which has higher final price, make consumers¡¦ final price prediction of group buying higher than the others under the best condition. On the other hand, it indicates that decreasing price curve, which has higher initial price, make consumers¡¦ final price prediction of group buying higher than the others under the worst condition. Consumers¡¦ internal reference prices are mainly influenced by market price information. When there are other posted-price stores to be chosen, consumers¡¦ internal reference price are higher averagely. In addition, the higher consumers¡¦ perceived transaction utilities are, the higher consumers¡¦ intentions to join group buying are. Finally, it shows consumers¡¦ intentions to join group buying have significant effect on their actual behavior.

Online group-buying is one of popular and innovative business models in the Internet age. Its essence is that the price will go down as the accumulated orders are increasing. A challenge is how to design the price curve, i.e., the relation between price and volume. Researchers have observed the consumer behavior phenomena of demand externalities, price-drop effect, and cycle ending effect in online group-buying transactions. All are related to the design of price curve. Therefore, if the price curve design can attract consumers and make profit as well, it will make the group-buying more successful. Based on above background, the purpose of this study is to explore the impact of price reduction frequency on consumers¡¦ perceived value and perceived transaction utility and then in turn on consumers¡¦ intention. In addition, this study also explores how the consumers¡¦ price sensitivity mediate the impact of price reduction frequency on consumers¡¦ perceived value and perceived transaction utility. A field experiment was done first. Then, an online experiment was designed and implemented based on the observation of the field experiment. The research result indicates that price reduction frequency will positively affect the consumers¡¦ perceived value and perceived transaction utility. Further, both consumers¡¦ perceived value and perceived transaction utility will affect the consumers¡¦ intention. In addition, the consumers¡¦ price sensitivity modulates the impact of price reduction frequency on consumers¡¦ perceived value and perceived transaction utility. Finally, the observations of field experiment and lab experiment demonstrate the phenomena of demand externalities, price-drop effect, and cycle ending effect proposed by previous researches.

To overcome the curse of dimensionality, dimension reduction is important and necessary for understanding the underlying phenomena in a variety of fields. Dimension reduction is the transformation of high-dimensional data into a meaningful representation in the low-dimensional space. It can be further classified into feature selection and feature extraction. In this thesis, which is composed of four projects, the first two focus on feature selection, and the last two concentrate on feature extraction. The content of the thesis is as follows. The first project presents several efficient methods for the sparse representation of a multiple measurement vector (MMV); some theoretical properties of the algorithms are also discussed. The second project introduces the NP-hardness problem for penalized likelihood estimators, including penalized least squares estimators, penalized least absolute deviation regression and penalized support vector machines. The third project focuses on the application of manifold learning in the analysis and prediction of 24-hour electricity price curves. The last project proposes a new hessian regularized nonlinear time-series model for prediction in time series.