Title :
JARTA — A Java library to model and fit Autoregressive-To-Anything processes
Author :
Uhlig, Tobias ; Rose, Oliver ; Rank, Sebastian
Author_Institution :
Dept. of Comput. Sci., Univ. der Bundeswehr Munchen, Neubiberg, Germany
Abstract :
JARTA is a Java library to model and fit Autoregressive-To-Anything (ARTA) processes. These processes are able to capture the dependency structure of a system, in contrast to commonly used models, that assume independently distributed random values. This study uses a simulation model of a warehouse to demonstrate the importance of capturing dependencies when modeling stochastic processes. Consequently there is a need for a suitable modeling approach. With JARTA we provide a modern software package to model processes with an appropriate dependency structure. Its two main goals are providing a clean code base for integration in other projects and high transparency for educational purposes. To support these goals JARTA is published under an open source license at http://sourceforge.net/projects/jarta/.
Keywords :
Java; autoregressive processes; mathematics computing; modelling; public domain software; software libraries; software packages; ARTA processes; JARTA; Java library; autoregressive-to-anything processes; open source license; software package; stochastic process modeling; system dependency structure; warehouse simulation model; Correlation; Distribution functions; Java; Libraries; Mathematical model; Stochastic processes; Unified modeling language;
Conference_Titel :
Simulation Conference (WSC), 2013 Winter
Conference_Location :
Washington, DC
Print_ISBN :
978-1-4799-2077-8
DOI :
10.1109/WSC.2013.6721508