by **Joe22c.** » Sat May 18, 2013 11:10 pm

The idea of one number appearing briefly before being replaced with another was a pretty cool touch. It reminds me of modeling software (e.g., MPlus) that is capable of performing more advanced statistic techniques.

Individuals who have taken an undergraduate statistics course (even advanced statistics courses taken in 3rd or 4th year) are typically only trained in basic "traditional" statistics based off Ordinary Least Squares (e.g., ANOVA, Multiple Regression) estimation wherein the "best" parameter estimate is the one that minimizes the error residual between each observed value and each predicted value.

A newer family of statistics are the "modeling" techniques based around Maximum Likelihood estimation which works kind of backwards - it's kind of hard to explain. While you can perform traditional statistics by hand, the more contemporary family of stats is so computationally demanding that you need computer programs to handle them. The process is "iterative" in the sense that for each "paramater" (i.e., statement) a value is provided, checked, re-provided over and over and over again until the numbers stop changing (or stop changing much). If the data you supply is non-sensical, the parameter estimates never stabilize (we say that the matrix does not converge) and it's one of the most frustrating things in the world ever.

This just reminded me of that. Of course, in this case, instead of 1,000 iterations, we only saw 1 (or 2, depending on how you count it) so I guess my analogy isn't apt. But anyways, I thought it was amusing.

Also - Question: Why doesn't Parson ask the bracer what the probability is that the scroll will do "______"? E.g., "What is the probability that this scroll will actually return me to Earth?" Seems like a worthwhile thing to have asked this whole time...