• Journal of the Korean Statistical Society
• /
• v.29 no.3
• /
• pp.337-350
• /
• 2000

This paper suggests a new diagnostic measure and a stopping rule for detecting influential observations in multiple discriminant analysis (MDA). It is developed from a Bayesian point of view using a default Bayes factor obtained from the fractional Bayes factor methodology. The Bayes factor is taken as a discriminatory information in MDA. It is shown that the effect of an observation over the discriminatory information is fully explained by the diagnostic measure. Based on the measure, we suggest a stopping rule for detecting influential observations in a given training sample. As a tool for interpreting the measure a graphical method is sued. Performance of the method is used. Performance of the method is examined through two illustrative examples.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.45 no.7
• /
• pp.53-61
• /
• 2008

Consider a typical wireless sensor network in which stem nodes form the backbone network of mesh topology while each stem node together with leaf nodes in its vicinity forms a subnetwork of star topology. In such a wireless sensor network, we must heed the following when we design a MAC scheme supporting the packet delivery from a leaf node to a stem node. First, leaf nodes are usually battery-powered and it is difficult to change or recharge their batteries. Secondly, a wireless sensor network is often deployed to collect and update data periodically. Late delivery of a data segment by a sensor node causes the sink node to defer data processing and the data segment itself to be obsolete. Thirdly, extensive signaling is extremely limited and complex computation is hardly supported. Taking account of these facts, a MAC scheme must be able to save energy and support timeliness in packet delivery while being simple and robust as well. In this paper, we propose a version of ALOHA as a MAC scheme for a wireless sensor network. While conserving the simplicity and robustness of the original version of ALOHA, the proposed version of ALOHA possesses a distinctive feature that a sensor node decides between stop and continuation prior to each delivery attempt for a packet. Such a decision needs a stopping rule and we suggest a Bayes stopping rule. Note that a Bayes stopping rule minimizes the Bayes risk which reflects the energy, timeliness and throughput losses. Also, a Bayes stopping rule is practical since a sensor node makes a decision only using its own history of delivery attempt results and the prior information about the failure in delivery attempt. Numerical examples confirm that the proposed version of ALOHA employing a Bayes stopping rule is a useful MAC scheme in the severe environment of wireless sensor network.

• Journal of the Korean Statistical Society
• /
• v.10
• /
• pp.128-139
• /
• 1981

The Bayesian sequential estimation problem for k parameters exponential families is considered using loss related to the Fisher information. Tractable expressions for the Bayes estimator and the posterior expected loss are found, and the myopic or one-step-ahead stopping rule is defined. Sufficient conditions are given for optimality of the myopic procedure, and the myopic procedure is shown to be asymptotically optimal in all cases considered.