Making decisions over multivariate time series is an important topic which has gained significant interest in the past decade. A time series is a sequence of data points which are measured and ordered over uniform time intervals. A multivariate time series is a set of multiple, related time series in a particular domain in which domain experts utilize multivariate time series to make a vital decision. Through studying multivariate time series, specialists are able to understand problems of events from different perspectives within particular domains. Identification and detection of those significant events over multivariate time series can lead to a better decision-making and actionable recommendations. Existing approaches to identifying and detecting significant events and delivering recommendations can be roughly divided into two categories: domain-knowledge-based and formal-learning-based. The former relies on solely domain experts' knowledge. Based on their knowledge and experience, domain experts can determine the conditions and the designed parameters to detect the events of interest and then deliver appropriate actions. However, those parameters are not always accurate. In addition, the parameters are static, but the problem that we deal with is often dynamic in nature. Thus the domain-knowledge-based approach lacks a formal mathematical foundation to dynamically learn parameters to meet the need of the changing environment. The latter approach, formal-learning-based, is to utilize formal learning methods such as non-linear logistic regression models. The logistic regression models are used to predict the occurrence of an event by learning parametric coefficients of the logistic distribution function of the explanatory variables. The challenge using formal learning methods to support decision-making is that they do not always produce satisfactory results, as they are not incorporated with domain experts' knowledge into their learning processes. Clearly, both approaches, domain-knowledge-based and formal-learning-based, do not take advantage of each other to learn optimal decision parameters, which are used to detect the events and then make better actionable recommendations. To support such an event-based decision-making and recommendation, I proposed a Web-Mashup Application Service Framework for Multivariate Time Series Analytics (MTSA). It is an integrated framework to support the MTSA service development and implementation, including parametric model definitions, query formulation, parameter learning, data monitoring, decision recommendations, and model evaluations. Domain experts could use the framework to develop and implement their web-based decision-making applications on the Internet. More specifically, the technical contributions of my dissertation include (1) the MTSA data model and query language to support the development and implementation of those MTSA services; (2) the hybrid-based mathematical models and computational algorithms that combine the strengths of both domain-knowledge-based and formal-learning-based approaches to learn decision parameters over multivariate time series; and (3) the experimental case studies to solve the real-world problems in two different domains, i.e., the stock markets and the electric power microgrids, to evaluate the models and algorithms, which are designed for the learning, monitoring, and recommendation services. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]