Timeline for Forcing a set of numbers to a gaussian bell-curve
Current License: CC BY-SA 3.0
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 23, 2017 at 12:39 | history | edited | CommunityBot |
replaced http://stackoverflow.com/ with https://stackoverflow.com/
|
|
| Dec 13, 2012 at 15:07 | history | edited | gung - Reinstate Monica | CC BY-SA 3.0 |
light editing & formatting for better readability; removed thanks
|
| Jan 10, 2011 at 20:38 | comment | added | Wayne | reading your last comment, I wonder what approach is best: give, e.g. 1 point for each loan paid off, other points for other things, then try to tame the enormous score at the end, or perhaps do something more meaningful with each component that makes up the final score? For example, if you give points for loans paid off you could calculate your Loans Paid Off (LPO) points as log (1 + NumberOfLoansPaidOff). So someone with zero loans paid off gets an LPO of 0, someone with 3 loans paid off gets an LPO of 1.1, while someone with 100 loans paid off gets an LPO of 4.6. Cap LPO at 5. | |
| Jan 1, 2011 at 0:00 | vote | accept | NealWalters | ||
| Dec 31, 2010 at 18:58 | answer | added | babelproofreader | timeline score: 4 | |
| Dec 31, 2010 at 17:47 | answer | added | whuber♦ | timeline score: 13 | |
| Dec 31, 2010 at 15:03 | history | edited | NealWalters | CC BY-SA 2.5 |
edited body
|
| Dec 31, 2010 at 14:59 | comment | added | NealWalters | @onestop - see my link to previouspost. In school, I remember of hearing of teachers who graded "on the curve". So no matter what score you got on a test, only a certain percentage of people would get each grade. I have no idea how credit scores work, but that's very similar to what I'm simulating. I give people points for example when they pay off a loan. But some people will pay off dozens of loans. I want to condense my wild points into a set of normalized points. | |
| Dec 31, 2010 at 11:37 | comment | added | onestop | Welcome to CrossValidated, NealWalters. You may find that a common response to asking a statistician "How do i do this" is "Why do you want to do that?" | |
| Dec 31, 2010 at 11:22 | history | edited | onestop | CC BY-SA 2.5 |
add a line break to fix the displayed numbering
|
| Dec 31, 2010 at 7:43 | comment | added | mpiktas | If your set changes time from time it is impossible to fit it to the same bell-curve. Suppose you have the set $A=\{1,2,3,4\}$, so 4 should be 800, but if a new observation 5 arrives, then it should become 800. | |
| Dec 31, 2010 at 5:49 | history | asked | NealWalters | CC BY-SA 2.5 |