NIST StRD Data Archive
Main
Univ Page
NumAcc4
Dataset
Additional Information

Dataset
Name:

NumAcc4

Procedure: Univariate Summary Statistics
Certification Method & Definitions

Data: 1 Response Variable (y)
1001 Observations
Higher Level of Difficulty
Observed Data

Model: 3 Parameters (mu, sigma, rho1)

y(i) = mu + e(i)


This generated/fabricated dataset consists of 1001 9-digit floating-point values: a single 10000000.2, followed by 500 pairings of 10000000.1 and 10000000.3. By construction, this data set has sample mean = 10000000.2 (exact); sample standard deviation = .1 (exact); and sample autocorrelation coef. = -0.999 (exact). The construction was carried out based on considerations described by Simon, Stephen D. and Lesage, James P. (1989): Assessing the Accuracy of ANOVA Caluclations in Statistical Software", Computational Statistics & data Analysis, 8, pp. 325-332.