Sequential Estimation of Quantal Response Curves
Generate an AI Snapshot to get a quick, structured summary of this paper.
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
Abstract
SUMMARY We suppose that it is required to estimate percentage points of a quantal response curve, and efficient use of observations is important. Strategies currently available are critically examined, including the Robbins-Monro process and Dixon and Mood’s Up and Down method. Some new alternative strategies are suggested and studied using empirical sampling on an electronic computer. The procedure recommended for estimation of 50 per cent. points has advantages of ease of operation and estimation, besides high efficiency; studies on the estimation of other percentage points were less successful. A sketch is given of problems demanding further study.