Publications: April 2011 Archives

Forecasting of time series data is crucial for decision-making processes in many domains as it allows the prediction of future behavior. In this context, a model is fit to the observed data points of the time series by estimating the model parameters. The computed parameters are then utilized to forecast future points in time. Existing approaches integrate forecasting into traditional relational query processing, where a forecast query requests the creation of a forecast model. Models of continued interest should be deployed only once and used many times afterwards. This however leads to additional maintenance costs as models need to be kept up-to-date. Costs can be reduced by choosing a well-defined subset of models and answering queries using derivation schemes. In contrast to materialized view selection, model selection opens a whole new problem area as results are approximate. A derivation schema might increase or decrease the accuracy of a forecast query. Thus, a two-dimensional optimization problem of minimizing the model cost and model usage error is introduced in this paper. Our solution consists of a greedy enumeration approach that empirically evaluates different configurations of forecast models. In our experimental evaluation, with data sets from different domains, we show the superiority of our approach over traditional approaches from forecasting literature.

Continuous balancing of energy demand and supply is a fundamental prerequisite for the stability and efficiency of energy grids. This balancing task requires accurate forecasts of future electricity consumption and production at any point in time. For this purpose, database systems need to be able to rapidly process forecasting queries and to provide accurate results in short time frames. However, time series from the electricity domain pose the challenge that measurements are constantly appended to the time series. Using a naive maintenance approach for such evolving time series would mean a re-estimation of the employed mathematical forecast model from scratch for each new measurement, which is very time consuming. We speed-up the forecast model maintenance by exploiting the particularities of electricity time series to reuse previously employed forecast models and their parameter combinations. These parameter combinations and information about the context in which they were valid are stored in a repository. We compare the current context with contexts from the repository to retrieve parameter combinations that were valid in similar contexts as starting points for further optimization. An evaluation shows that our approach improves the maintenance process especially for complex models by providing more accurate forecasts in less time than comparable estimation methods.