Timeline for Fake uniform random numbers: More evenly distributed than true uniform data
Current License: CC BY-SA 3.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 15, 2012 at 13:48 | comment | added | Peter Flom | OK, but, in the example in the question, he wants to place treasure on a map in a game. That won't involve inference or moments or anything of the sort. I admit my method wouldn't be good for a lot of purposes, but I think it matches up with the example. Of course, maybe the example isn't really what he wants.... Maybe he wants something more formal, in which case all the other answers should be looked at. | |
| Oct 15, 2012 at 13:21 | comment | added | whuber♦ | Let me state my objection a little more clearly, Peter: when you remove and/or adjust pseudorandom values in an ad hoc way in order to approximate some desired property, such as lack of clustering, it is difficult to assure that the resulting sequences have any desirable properties. With your method, for instance, could you even tell us what the first moment of the resulting process would be? (That is, can you even assure us that the intensity is uniform?) What about the second moment? Usually these constitute the minimum information needed to use the sequences effectively for inference. | |
| Oct 14, 2012 at 19:43 | comment | added | Has QUIT--Anony-Mousse | I had thought about that, but it doesn't work well with the incremental part. As long as I have a fixed (or upper bounded) requirement of samples, I can reject and retry. But a "constructive" method would be nicer. Plus, in particular for small sample sizes, the tests might not be very strong. After all, they usually try to avoid false rejects, and the data was uniformly distributed. | |
| Oct 14, 2012 at 18:18 | comment | added | Peter Flom | @whuber He doesn't seem to be interested in that, although I might be wrong. | |
| Oct 14, 2012 at 17:21 | comment | added | John | The OP seems most concerned with small sample sizes. This would suggest he doesn't need to care about the whole distribution. Suppose you have a set of coordinates, you generate another and then calculate the euclidean distance with respect to all the others. If the smallest distance is below some threshold, throw the number out and generate a new one. I think Peter's solution works fine. | |
| Oct 14, 2012 at 16:18 | comment | added | whuber♦ | One of the (many) problems with this approach is it's very hard to characterize the resulting distribution. | |
| Oct 14, 2012 at 16:10 | history | answered | Peter Flom | CC BY-SA 3.0 |