Glossary entry (derived from question below)
Spanish term or phrase:
regresiones logisticas binarias y policotomicas
English translation:
binary and multinomial logistic regressions
Added to glossary by
Caroline Clarke
Jun 10, 2016 13:55
7 yrs ago
3 viewers *
Spanish term
regresiones logisticas binarias y policotomicas
Spanish to English
Science
Science (general)
statistics
This is in a Mexican medical document, describing statistical analysis of the results of a survey for a study. The sentence is, "Se obtendran las prevalencias de consumo, las tendencias y los intervalos de confianza a traves del programa STATA y se evaluaran los factores asociados al consumo a traves de regresiones logisticas binarias y policotomicas. I have everything except the last word. (sorry, this won't let me use accents!)
I'm thinking along the lines of multivariant logistical, multiple category... I even found "multichotomous" but I can't understand the definition. My expertise is medical translation, with basic statistics terminology!
I'm thinking along the lines of multivariant logistical, multiple category... I even found "multichotomous" but I can't understand the definition. My expertise is medical translation, with basic statistics terminology!
Proposed translations
(English)
4 +4 | binary and multinomial logistic regressions | Caroline Clarke |
3 +2 | polytomous | Dr Jane Marshall |
4 +1 | binary and polychotomous logistic regressions | philgoddard |
References
Wikipedia says... | Anne Schulz |
Change log
Jun 13, 2016 09:31: Caroline Clarke Created KOG entry
Proposed translations
+4
1 hr
Selected
binary and multinomial logistic regressions
I think that "multinomial" is the word you want.
--------------------------------------------------
Note added at 1 day38 mins (2016-06-11 14:34:25 GMT)
--------------------------------------------------
Bog-standard logistic regression is binary, i.e. the dependent variable is binary (can take one of two outcomes, e.g. doing a regression analysis to predict whether a patient is alive or dead after x amount of time), and the extension of this is called multinomial, when the dependent variable has 3 or more categories (e.g. doing a regression analysis to predict whether a patient has mild, moderate or severe depression in x amount of time).
I have never seen the 3+ version (i.e. more than binary/binomial) called anything other than multinomial.
Thanks for the discussion, everyone! :)
--------------------------------------------------
Note added at 1 day38 mins (2016-06-11 14:34:25 GMT)
--------------------------------------------------
Bog-standard logistic regression is binary, i.e. the dependent variable is binary (can take one of two outcomes, e.g. doing a regression analysis to predict whether a patient is alive or dead after x amount of time), and the extension of this is called multinomial, when the dependent variable has 3 or more categories (e.g. doing a regression analysis to predict whether a patient has mild, moderate or severe depression in x amount of time).
I have never seen the 3+ version (i.e. more than binary/binomial) called anything other than multinomial.
Thanks for the discussion, everyone! :)
Peer comment(s):
agree |
Francois Boye
: The two concepts (multinomial and polychotomous) are the same// I am sorry, the concept of polychotomous variable does exist! See the attachment: http://www.statisticshowto.com/polychotomous-variable/
1 hr
|
Thanks for the discussion, everyone!
|
|
agree |
Taña Dalglish
: You deserve the agree. I don't understand Francois and it does not merit a neutral, as you made no reference to both terms. https://onlinecourses.science.psu.edu/stat504/node/172 (polychotomous according to this does/should not exist). Regards Caroline
2 hrs
|
Thanks for the discussion, everyone!
|
|
agree |
Robert Carter
: Not my area either, but this appears to be correct too.
9 hrs
|
Thanks for the discussion, everyone!
|
|
agree |
Anne Schulz
: 'multinomial' is more commonly used than 'polytomous', IMO
16 hrs
|
Thanks for the discussion, everyone!
|
4 KudoZ points awarded for this answer.
Comment: "Thanks to everyone for the lesson in synonyms. I'm going with "binomial and multinomial" logistic regressions. It flows better."
+2
17 mins
polytomous
I wonder if it could be a mistake for polytomous logistic regression?
--------------------------------------------------
Note added at 21 mins (2016-06-10 14:17:13 GMT)
--------------------------------------------------
ie, they did binary/binomial logistic regression and polytomous logistic regression. I am not a big fan of statistics either!
--------------------------------------------------
Note added at 21 mins (2016-06-10 14:17:13 GMT)
--------------------------------------------------
ie, they did binary/binomial logistic regression and polytomous logistic regression. I am not a big fan of statistics either!
Peer comment(s):
agree |
Robert Carter
: Agree with your addendum. Well spotted.
10 hrs
|
agree |
Anne Schulz
17 hrs
|
+1
19 mins
binary and polychotomous logistic regressions
I'm not going to bother with a long explanation, because this is even further out of my field than it is yours!
"Logistical regression" does get hits, but "logistic" gets far more, so I suspect the former may be a widespread mistake.
--------------------------------------------------
Note added at 22 mins (2016-06-10 14:18:14 GMT)
--------------------------------------------------
Theotry and method: polychotomous regression
http://amstat.tandfonline.com/doi/abs/10.1080/01621459.1997....
"Logistical regression" does get hits, but "logistic" gets far more, so I suspect the former may be a widespread mistake.
--------------------------------------------------
Note added at 22 mins (2016-06-10 14:18:14 GMT)
--------------------------------------------------
Theotry and method: polychotomous regression
http://amstat.tandfonline.com/doi/abs/10.1080/01621459.1997....
Peer comment(s):
neutral |
Robert Carter
: Sorry Phil, as Taña has mentioned above, polychotomous doesn't exist.//Perhaps I should have said "shouldn't exist" :-)//Be careful with Google hits, there are actually only 88, not 54,000.
10 hrs
|
Of course it does! It's in Webster's, and it gets 54,000 hits. The Spanish word for "multinomial" is "multinomial".
|
|
agree |
Francois Boye
: I cannot dispute the existence of the article attached by Phil. However, I agree that the concept of multinomial variable is by far more used in the bio/econometric literature and softwares.
10 hrs
|
Reference comments
18 hrs
Reference:
Wikipedia says...
In statistics, multinomial logistic regression is a classification method that generalizes logistic regression to multiclass problems, i.e. with more than two possible discrete outcomes. That is, it is a model that is used to predict the probabilities of the different possible outcomes of a categorically distributed dependent variable, given a set of independent variables (which may be real-valued, binary-valued, categorical-valued, etc.).
Multinomial logistic regression is known by a variety of other names, including polytomous LR, multiclass LR, softmax regression, multinomial logit, maximum entropy (MaxEnt) classifier, conditional maximum entropy model.
https://en.wikipedia.org/wiki/Multinomial_logistic_regressio...
Multinomial logistic regression is known by a variety of other names, including polytomous LR, multiclass LR, softmax regression, multinomial logit, maximum entropy (MaxEnt) classifier, conditional maximum entropy model.
https://en.wikipedia.org/wiki/Multinomial_logistic_regressio...
Discussion
There does, however, appear to be some difference between the two words, if we are to believe the Oxford:
polychotomy
NOUN
Division into more than two parts or groups; especially ( Taxonomy ) the division of a taxonomic grouping into more than two subgroups; an instance of this.
http://www.oxforddictionaries.com/definition/english/polytom...
cf
polytomy
NOUN
1 Logic and Statistics . A division or classification into several (usually more than three) groups or subsets; (in later use also) the assignment of three or more possible values to a categorical variable.
2 Biology and Taxonomy . Division into more than two branches at the same point or node; the occurrence of such branching.
http://www.oxforddictionaries.com/definition/english/polytom...
So, perhaps "polytomous" is the word that should be used, in any case, and "polychotomous", a solecism originally coined in taxonomy, has mistakenly gained currency in statistics literature.
The term dichotomy is from the Greek language διχοτομία dichotomía "dividing in two" from δίχα dícha "in two, asunder" and τομή tomḗ "a cutting, incision".
https://en.wikipedia.org/wiki/Dichotomy
8.1 - Polytomous (Multinomial) Logistic Regression | STAT 504
https://onlinecourses.science.psu.edu/stat504/node/172
We have already learned about binary logistic regression, where the ... (Note: The word polychotomous is sometimes used, but this word does not exist!).
Logistic Regression Models for Multinomial and Ordinal Variables
www.theanalysisfactor.com/logistic-regression-models-for-mu...
The multinomial (a.k.a. polytomous) logistic regression model is a simple extension of the binomial logistic regression model. ... In multinomial logistic regression the dependent variable is dummy coded into multiple 1/0 variables. ... The mulitnomial logistic regression then estimates a ...
It could not be further from my working field, so I leave it at that. Regards.