level.ventil causing inconsistent MCA results

[Gilad Brandes]

Hi everyone,

I recently posted a question about HCPC giving out different results with each run, but I've come to realize that the problem is not with HCPC but with MCA, and particularly with level.ventil:

One of my variables contains several very infrequent categories, and ventilating these categories seems to be what's causing the problem. When level.ventil=0 the MCA results are consistent every time, but when level.ventil>0 I get different results with each run, even though the code remains exactly the same.

I guess I don't fully understand what ventilation actually does... Any suggestions?

[montoy...@gmail.com]

Hello,

Another solution is to merge the category which has little frequency with its closest category. For example, if you have a category "Over 75 years old" with few individuals, you merge it with the category "Between 60 and 74 years old", thus creating a new category "Between 60 and over 75 years old" which will be more frequent. You can then keep the old categories as additional. In Husson/Lê/Pagès, "Analyse de données avec R"  2016,  Section "3-7-1 Complement" -> "Prise en compte des modalités rares" p.149-150

Victor

[Gilad Brandes]

Thank you Victor!

But I don't think that's the right solution in my case, since these categories cannot be merged with each other in any meaningful way. 

[Francois Husson]

ventilation means that individuals that take an infrequent category will be affected to another category randomly. That is why the results are not the same from one run to another.
FH

--
Vous recevez ce message, car vous êtes abonné au groupe Google Groupes "FactoMineR users".
Pour vous désabonner de ce groupe et ne plus recevoir d'e-mails le concernant, envoyez un e-mail à l'adresse factominer-use...@googlegroups.com.
Cette discussion peut être lue sur le Web à l'adresse https://groups.google.com/d/msgid/factominer-users/cf16229c-b101-4505-9bbe-39cdc7b2d5afn%40googlegroups.com.

--
Francois Husson
Department Statistics & Computer science
AGROCAMPUS OUEST
65 rue de St-Brieuc - 35042 RENNES
Tel: +33 2 23 48 58 86
https://husson.github.io

[Gilad Brandes]

Thank you, that certainly explains it. 

‫בתאריך יום ג׳, 3 בנוב׳ 2020 ב-12:20 מאת ‪Francois Husson‬‏ <‪francoi...@agrocampus-ouest.fr‬‏>:‬

Vous recevez ce message, car vous êtes abonné à un sujet dans le groupe Google Groupes "FactoMineR users".
Pour vous désabonner de ce sujet, visitez le site https://groups.google.com/d/topic/factominer-users/T-C49RB35aI/unsubscribe.
Pour vous désabonner de ce groupe et de tous ses sujets, envoyez un e-mail à l'adresse factominer-use...@googlegroups.com.
Cette discussion peut être lue sur le Web à l'adresse https://groups.google.com/d/msgid/factominer-users/d6c7d9b8-6e25-07c7-861d-53c006173343%40agrocampus-ouest.fr.