Computes, under Gaussian item assumption: point estimates, analytically; and confidence intervals, by Monte Carlo simulation.

Many, N, samples of T (trucks) weights, which are sums of n items (bags), are simulated. These N vectors of size T have their own averages and variances, whose behavior is studied.

The (particular) case use of single, i.e., equal, sample sizes is an attempt to validate the general case (elsewhere) of unequal size samples. By simulation, it leads to expected Student's and chi-squared distributions.

With (T n μ σ) = (9 1 25. 1.), it gives CIs (24.35 25.65, 0.27 2.19); with (9 5 25. 1.), it gives CIs (24.71 25.29, 0.27 2.19).

Draws plots of the behavior of locations and variabilities.

^* 'Month' stands for a certain period where T "trucks" carried n "bags" each (typical setting).

• μ ∈ x^– ± t_{α⁄ 2, n-1} . s ⁄ √(n)  • (n - 1)s² / χR ≤ σ² ≤ (n - 1)s² / χL, with χL = χ²_{α⁄ 2, n-1}, and χR = χ²_{1-α⁄ 2, n-1} .

• (NIST) t Distribution; Confidence interval for σ

• Confidence intervals based on normal data (MIT) • Confidence Intervals (J. Gehrman, California State University, Sacramento), e.g..

• Montgomery, Douglas C., 2013, "Introduction to Statistical Quality Control", 7.^th ed., Wiley, Hoboken, NJ (USA). ISBN 78-1-118-14681-1, pp 122–128.

• 1888-06-24: Darmois, George († 1960-01-03, 71 yrs.).