Lag Sequential Analysis Using Pronet: Effects of Series Length and Noise
Preston A. Kiekel, Jennifer L. Winner · Proceedings of the Human Factors and Ergonomics Society Annual Meeting · 2005
Procedural Networks (ProNet) is presented as a form of lag sequential analysis, but using the Pathfinder algorithm as its kernel. In general, Lag Sequential Analysis is a tool used to identify representative chains of events in a time series of event data. A simulation is used to compare the performance of ProNet, as sample size and noise are varied in the time series. Noise in the data leads to fewer hits, but does not affect false alarm rate. Increased sample size leads to fewer false alarms and more hits. The method is defined and developed, and future research is discussed.