Skip to content

Generalized Linear Model

4 messages · Megan, Alexander Engelhardt, Kehl Dániel +1 more

#
Hello,

I am trying to run a generalized linear model but do not know where to
begin. I have attached my data to R but do not know where to go from there.
I have two independent variables (each has two factors associated with them)
and two dependent variables, each with either a yes/no response which I've
valued either 0 or 1 in the data set. Any input would be greatly
appreciated. 

--
View this message in context: http://r.789695.n4.nabble.com/Generalized-Linear-Model-tp3473924p3473924.html
Sent from the R help mailing list archive at Nabble.com.
#
Am 25.04.2011 21:28, schrieb Megan:
What do you mean by this? You have two input variables, who are binary, 
meaning yes/no (or male/female, high/low, ...) variables?
If your dependent variable is binary, you might want to google for 
"logistic regression" (this belongs to generalized linear models). The 
R-function who handles this is glm(), with the parameter family=binomial().

Have fun,
  Alex
#
Hi!

Try to read about the glm function, type:

?glm

in your R editor. It looks like you have contingency tables, maybe a 
loglin model would be good to start with.

D

2011-04-25 12:28 keltez?ssel, Megan ?rta:
1 day later
#
Because you have two dependent variables, you'll want to to use a
multivariate logit. mlogit does this, but I don't know the syntax off
hand.

If you just wanted to look at one dependent variable, it would be the
following (which Alex said)

glm(y~x1*x2,family='binomial')
On Mon, Apr 25, 2011 at 3:28 PM, Megan <aforkonaplate at hotmail.com> wrote: