Predictor Sort Confidence Interval Simulation

This program performs a simulation to test how well corrected confidence intervals on means based on ANOVAs and ANOCOVs perform under predictor sort sampling. It also estimates coverages for for confidence intervals based on uncorrected blocked and unblocked ANOVAs.

This work is discussed in two papers: Verrill (1999), "When Good Confidence Intervals Go Bad: Predictor Sort Experiments and ANOVA," The American Statistician, 53(1), 38-42 (pdf) and Verrill and Kretschmann (2017a), "A Reminder about Potentially Serious Problems with a Type of Blocked ANOVA Analysis," USDA Forest Products Laboratory research paper FPL-RP-683 (pdf).

Links to the FORTRAN source code for a version of this program can be found here.

How many factors are there?
(e.g., 3 for a three-way ANOVA)

How many levels are there?
(e.g., 3 2 2 for a 3x2x2 ANOVA.)

How many replicates are there per factor combination?
(e.g., to have 5 replicates in a 3x2x2 ANOVA, a total of 5x3x2x2 specimens would be required.)

For confidence intervals on which factor do you want to perform a simulation?
(e.g., if the experiment were a 5x2x4 ANOVA and you were interested in a confidence interval for one of the 4 levels of factor three, you would choose 3.)

What is the correlation value?
(strictly between -1.0 and 1.0)

How many trials do you want in the simulation?
(Between 1 and 10000. Please note that the simulation time increases roughly as numtr x n x ln(n) where numtr is the number of trials and n is the sample size. Please try to avoid simulations that take excessive time. A 2x5 with 5 replicates and 4000 trials takes about 20 seconds.)

What is the starting value for the random number generator?
(It must be a positive integer less than 1000000000.)

The "random" numbers that are generated are exactly determined by the starting value. Thus, if you want a different set of "random" numbers, you must supply a different starting value.

What name do you want for your results file?

The name should be unique to you to prevent the file from being accidentally overwritten by another user.

Since you will see an html document that presents the results as soon as you execute the program, you may not want an ascii copy. However, we do provide this option.

The ascii file will be written in the pub/tt anonymous ftp directory. If a file by the same name already exists in the directory, the new results will be appended to the existing file. Check the anonymous ftp link for directions about retrieving the file.

If ftp does not run on your machine, e-mail me at and ask me to e-mail the results file to you. Do this quickly as the pub/tt directory is cleaned weekly.

For further information, please contact Steve Verrill at or 608-231-9375.
Last modified on 1/12/17.

As of last midnight, this page had been accessed times.