Timeline for What is the inverse normal transformation (INT) and what are the reasons behind using it?
Current License: CC BY-SA 4.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 18, 2022 at 20:20 | comment | added | Nick Cox | The context here is linear regression in which an ideal condition is normal (Gaussian) distribution of errors (loosely, residuals). In contrast, invocation here of an inverse Gaussian is a complete red herring. FWIW, the suggestion appears to confuse modeling a series in time after a dramatic event with fitting a probability distribution; also the inverse Gaussian is usually a two-parameter distribution. | |
| Sep 18, 2022 at 13:40 | comment | added | user78229 | @whuber Thanks for your comment. Having read this literature, I stand by my remark. | |
| Sep 16, 2022 at 13:13 | comment | added | Nick Cox | Trivially, INT is best avoided as an abbreviation here, as in many languages and environments it is the name of a function that truncates to integer. | |
| Sep 16, 2022 at 12:55 | comment | added | whuber♦ | @Mike This is not the relevant meaning of "inverse." ("Making it appear more normally distributed" gives a strong clue as to the intention.) Researching the IGD will just take the OP on a fruitless and likely frustrating chase. The relevant sense is that of applying the inverse of the standard Normal CDF $\Phi$ to a suitably centered and scaled version of the data. | |
| Sep 16, 2022 at 11:48 | answer | added | Maarten Buis | timeline score: 4 | |
| Sep 16, 2022 at 11:14 | comment | added | user78229 | To make it simple, use these keywords inverse gaussian distribution in a Google Books search, multiple items will be returned. Basically, the IGD is a one parameter distribution which can be used in modeling explosive natural phenomena such as the flow of water from a bursting dam or sales upon release of an 80s Michael Jackson hit record. | |
| S Sep 16, 2022 at 11:09 | review | First questions | |||
| Sep 16, 2022 at 11:16 | |||||
| S Sep 16, 2022 at 11:09 | history | asked | 7-x | CC BY-SA 4.0 |