Actually, they can be. However, the genesis of the numbers is the CDC. The origin of your refutation comes from your imagination. Your objections are based on nothing.
No. Random numbers are not data. It does not matter if the CDC generates them or whether you generate them.
I do not have to prove a negative. Attempting to force a negative proof is a fallacy. YOU have to show validity of any data. I have a higher standard than most.
I do not accept any data unless I:
* know who collected it and when.
* know the boundary of collection.
* know when it was collected.
* know the method by which it was collected, including any instrumentation used.
* know the calibration method used for any such instruments.
* have the raw data published, along with all of these other requirements. You cannot use cooked data as raw data.
People manufacture numbers all the time. They are not data. The government is especially good at this. Quoting them means nothing.
In addition:
* if a summary is used, the variance (sometimes called the variation source) must be declared and justified.
* selection MUST be by randN. That selection MUST be from raw data.
* normalization MUST be by paired randR.
* margin error MUST be calculated and accompany the summary.
* no statistical summary is a crystal ball. It cannot predict.
These are my requirements. They are not out of line. Meet them or I cannot accept your so-called 'data'.
For more details, see the forum at my sig under the area known as The Data Mine.
Publishing any summary or statistic out of context is a special pleading fallacy or contextomy fallacy.