Timeline for How to transform a response variable with negative values?
Current License: CC BY-SA 3.0
15 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 5, 2017 at 2:33 | comment | added | MSS | @Glen-b, the third set of plots looks good to me, I was asking because of the output, but it may be an issue about not having a proportion at the end with the Poisson model, once back-transformed the units of the predicted values are eggs, not 0-1. Or it is possible? Thanks! | |
| May 3, 2017 at 11:22 | comment | added | Glen_b | While beta regression can adapt to deal with 1's (see 0-1 inflated beta regression) beta regression doesn't make sense because the proportions are ratios of counts. Perhaps I am being naive but I don't see what you think the problem is with the plots at the bottom of your post. What's the issue there? (I am not saying that the model is right there ... I just don't necessarily see what you think the problem might be). What do you think it should look like if the model was correct? | |
| May 1, 2017 at 22:56 | vote | accept | MSS | ||
| May 1, 2017 at 22:48 | history | edited | MSS | CC BY-SA 3.0 |
added 187 characters in body
|
| May 1, 2017 at 15:58 | answer | added | Wayne | timeline score: 5 | |
| May 1, 2017 at 15:32 | history | edited | MSS | CC BY-SA 3.0 |
added 142 characters in body
|
| May 1, 2017 at 15:13 | history | edited | MSS | CC BY-SA 3.0 |
deleted 116 characters in body
|
| May 1, 2017 at 15:03 | history | edited | MSS | CC BY-SA 3.0 |
added 124 characters in body
|
| May 1, 2017 at 15:02 | comment | added | MSS | @Glen_b♦, the original response variable is the number of eggs in the nest at hatch divided by the total number of eggs found (i.e. the proportion that survived incubation period). Regarding other alternatives, Beta regression doesn't work in my case because I have values =1; I tried with Poisson and logistic regression with weights for the denominator (please see the plots on the update of my question - I shared only for binomial regression because both look similar). Can you recommend me another approach? I really don't know what else can I try. Thanks a lot for your time. | |
| Apr 30, 2017 at 23:42 | comment | added | Glen_b | [Linear regression doesn't generally make sense if you're trying to model proportions of things; among numerous other problems, you'll tend to end up with a model that predicts impossible values] | |
| Apr 30, 2017 at 23:35 | comment | added | Glen_b | You deal with that problem via incorporating number of eggs into a suitable model for the proportion -- not by (what seems to be) fairly arbitrarily mangling the data. Perhaps you could explain in more detail what your original variables are. You cannot transform discrete data across only a few values to normality. Since you can only move the data values around you'll always end up with at most five different values and the height of each spike will not change -- so there will always be only a few discrete values. | |
| Apr 29, 2017 at 20:13 | comment | added | MSS | Thanks for your comment @PtrZink it is because the response variable cannot be analyzed directly, the proportions will have different values depending on the number of eggs in the nest, I am following this paper, please look at page 325 for more detail (last paragraph on the left column): canuck.dnr.cornell.edu/research/pubs/pdf/age_effects.pdf | |
| Apr 29, 2017 at 19:51 | comment | added | PtrZlnk | You said: "(Y) is the difference between the proportion of eggs found in a given nest and the expected value of eggs regarding the mean value of the sample". Firstly: what is the point of subtracting sample mean from proportion? Secondly: maybe the strange pattern in your data is result of this strange operation? | |
| Apr 29, 2017 at 19:26 | history | edited | MSS | CC BY-SA 3.0 |
added 224 characters in body
|
| Apr 29, 2017 at 19:11 | history | asked | MSS | CC BY-SA 3.0 |