SHORTEST-LENGTH PREDICTION INTERVALS FOR FUTURE OUTCOMES


Nicholas A. Nechval 1, Konstantin N. Nechval 2, Edgars K. Vasermanis 1
1 University of Latvia, Department of Mathematical Statistics, Raina Blvd 19, LV-1050, Riga, Latvia,
e-mail: nechval@junik.lv
2 Transport and Telecommunication Institute, Department of Computer Science, Lomonosov Street 1, LV-1019, Riga, Latvia
e-mail: konstan@tsi.lv

ABSTRACT:In this paper, a technique is proposed for constructing shortest-length prediction intervals for future outcomes. The prediction intervals depend upon a previously available complete or type II censored sample from the same distribution belonging to invariant family. Both new-sample prediction (e.g., using data from a previous sample to make predictions, say, on the future failure time of a new unit) and within-sample prediction (e.g., predicting the number of future failures from a sample, based on early data from that sample) problems are considered. Prediction intervals are required as specifications on future life for components, as warranty limits for the future performance of a specified number of systems with standby units, and in various other applications. The purpose of this paper is to give a simple technique for deriving prediction intervals with a minimum length property and providing exactly the nominal coverage probability.

KEYWORDS: Future Outcomes, Prediction Intervals, Shortest Length, Technique for Constructing.

---

Back to HMS2003 Home Page