In a previous article* the propagation of round-off error was studied for one operation with approximate numbers. Extending the method to more than one operation is not difficult in principle, but the details of keeping track of various ranges of integration are troublesome, as mentioned by Inman [3], and Woodward [4]. This paper presents a geometric device capable of handling the intricacies involved for a wide class of problems. Problem: Given a random sample x1, X2, X3 and x4of size four drawn from the rectangular distribution
No takes yet. Share an insight, caveat, or question.
Locker et al. (1962) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: