[R] Plot for Binomial GLM

Mon Oct 4 21:08:00 CEST 2010

Hi,

Dennis was kind of enough to remind me that glm() can take a two
column matrix, which is probably what you did with deadalive.  He also
gave a rather elegant graphing solution using xyplot:

xyplot(Alive/20 ~ Dose, data = rat.toxic, groups = Sex, type = c('p', 'a'))

Josh

On Mon, Oct 4, 2010 at 8:23 AM, Joshua Wiley <jwiley.psych at gmail.com> wrote:
> On Mon, Oct 4, 2010 at 7:21 AM, klsk89 <karenklsk89 at yahoo.com> wrote:
>>
>> Hi i would like to use some graphs or tables to explore the data and make
>> some sensible guesses of what  to expect to see in a glm model to assess if
>> toxin concentration and sex have a relationship with the kill rate of rats.
>> But i cant seem to work it out as i have two predictor
>> variables~help?Thanks.:)
>
> What about xtabs?   For instance:
>
> xtabs(deadalive ~ Dose + Sex, data = rat.toxic)
>
> Regarding graphs, take a look at faceting in ggplot2 (or lattice).
> You can get something close to the 3 way table but in graphical form
> that way.  I am not sure if this is completely up and running yet, but
> I know there has been work linking ggobi with R.  I have seen a few
> demonstrations that looked quite promising, and it may work well for
> you to visualize three variables at once (and interactively).  Here is
> the link:  http://www.ggobi.org/rggobi/
>
>>
>> Here's my data.
>>
>>> rat.toxic<-read.table(file="Rats.csv",header=T,row.names=NULL,sep=",")
>>> attach(rat.toxic)
>      ^  why attach it?
>>> names(rat.toxic)
>> [1] "Dose"  "Sex"   "Dead"  "Alive"
>>> rat.toxic
>>   Dose Sex Dead Alive
>> 1    10   F    1    19
>> 2    10   M    0    20
>> 3    20   F    4    16
>> 4    20   M    4    16
>> 5    30   F    9    11
>> 6    30   M    8    12
>> 7    40   F   13     7
>> 8    40   M   13     7
>> 9    50   F   18     2
>> 10   50   M   17     3
>> 11   60   F   20     0
>> 12   60   M   16     4
>> 13   10   F    3    17
>> 14   10   M    1    19
>> 15   20   F    2    18
>> 16   20   M    2    18
>> 17   30   F   10    10
>> 18   30   M    8    12
>> 19   40   F   14     6
>> 20   40   M   12     8
>> 21   50   F   16     4
>> 22   50   M   13     7
>> 23   60   F   18     2
>> 24   60   M   16     4
>
> Please tell me that after this, you converted the counts of dead and
> alive into a single variable that had a 0 or 1 if dead and the
> opposite as alive before you used it as the dependent variable in your
> logistic regression.
>
>> glm2<-glm(deadalive~Dose*Sex,family=binomial,data=rat.toxic)
>>> anova(glm2,test="Chi")
>> Analysis of Deviance Table
>>
>> Model: binomial, link: logit
>>
>> Response: deadalive
>>
>> Terms added sequentially (first to last)
>>
>>
>>         Df Deviance Resid. Df Resid. Dev P(>|Chi|)
>> NULL                        23    225.455
>> Dose      1  202.366        22     23.090    <2e-16 ***
>> Sex       1    4.328        21     18.762    0.0375 *
>> Dose:Sex  1    1.149        20     17.613    0.2838
>> ---
>> Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
>>> summary(glm2)
>>
>> Call:
>> glm(formula = deadalive ~ Dose * Sex, family = binomial, data = rat.toxic)
>>
>> Deviance Residuals:
>>     Min        1Q    Median        3Q       Max
>> -1.82241  -0.85632   0.06675   0.61981   1.47874
>>
>> Coefficients:
>>            Estimate Std. Error z value Pr(>|z|)
>> (Intercept) -3.47939    0.46167  -7.537 4.83e-14 ***
>> Dose         0.10597    0.01286   8.243  < 2e-16 ***
>> SexM         0.15501    0.63974   0.242    0.809
>> Dose:SexM   -0.01821    0.01707  -1.067    0.286
>> ---
>> Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
>>
>> (Dispersion parameter for binomial family taken to be 1)
>>
>>    Null deviance: 225.455  on 23  degrees of freedom
>> Residual deviance:  17.613  on 20  degrees of freedom
>> AIC: 91.115
>>
>> Number of Fisher Scoring iterations: 4
>>
>>
>>
>>
>>
>>
>> --
>> View this message in context: http://r.789695.n4.nabble.com/Plot-for-Binomial-GLM-tp2954406p2954406.html
>> Sent from the R help mailing list archive at Nabble.com.
>>
>> ______________________________________________
>> R-help at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>>
>
>
>
> --
> Joshua Wiley
> Ph.D. Student, Health Psychology
> University of California, Los Angeles
> http://www.joshuawiley.com/