R help - May 2011 - matching by gender and age

Hello R gurus, I have a data set from which i have to extract the gender and
age matched rows from controls and disease group
disease<-paste(rep(c('y','n'),11))
gender<-paste(rep(c('m','f'),11))
mcp<-rnorm(700,1400)
age<-rnorm(32,34)

dat<-data.frame(disease=disease,sex=gender,Dr_age=age[1:22],MCP=mcp[1:22])

I have other categorical variables also to add to the matching. all the
posts I came across are matching for a single column from two
matrix/dataframe. How can i match for multiple variables when all the data
is in one single dataframe





--
View this message in context:
http://r.789695.n4.nabble.com/matching-by-gender-and-age-tp3552825p3552825.html
Sent from the R help mailing list archive at Nabble.com.

On May 26, 2011, at 11:19 AM, 1Rnwb wrote:
> Hello R gurus, I have a data set from which i have to extract the  
> gender and
> age matched rows from controls and disease group
You need to define what you mean by "age-matched". Your example  
creates a very narrow age range which further adds questions about  
what you are doing. You probably need to look at defining an age- 
category variable with cut() and proceeding from there.

> disease<-paste(rep(c('y','n'),11))
> gender<-paste(rep(c('m','f'),11))
> mcp<-rnorm(700,1400)
> age<-rnorm(32,34)
>
> dat<- 
> data.frame(disease=disease,sex=gender,Dr_age=age[1:22],MCP=mcp[1:22])
Use set.seed to generate a reproducible input and then tell us what  
your expected output should be.
>
> I have other categorical variables also to add to the matching. all  
> the
> posts I came across are matching for a single column from two
> matrix/dataframe. How can i match for multiple variables when all  
> the data
> is in one single dataframe
>

David Winsemius, MD
West Hartford, CT

On Jun 3, 2011, at 1:37 PM, 1Rnwb wrote:
> Thanks, for pointing out the package e1071, the example for  
> matchControls is
> exactly what I am looking for, however how can I add additional  
> factors to
> match for.
> library(e1071)
> Age.case <- 40 + 5 * rnorm(50)
> Age.cont <- 45 + 10 * rnorm(150)
> Age <- c(Age.case, Age.cont)
>
> Sex.case <- sample(c("M", "F"), 50, prob = c(.4,
.6), replace = TRUE)
> Sex.cont <- sample(c("M", "F"), 150, prob = c(.6,
.4), replace = TRUE)
> Sex <- as.factor(c(Sex.case, Sex.cont))
>
> casecont <- as.factor(c(rep("case", 50), rep("cont",
150)))
> risk.case<-sample(c("H", "M", "L"), 50,
prob = c(.4,.4, .2), replace
> = TRUE)
> risk.cont<-sample(c("H", "M", "L"), 150,
prob = c(.3,.3, .4),
> replace > TRUE)
> risk<-as.factor(c(risk.case, risk.cont))
>
>
> ## now look at the group properties:
> boxplot(Age ~ casecont)
> barplot(table(Sex, casecont), beside = TRUE)
> barplot(table(Sex, risk), beside = TRUE)
>
> m <- matchControls(casecont ~ Sex + Age+risk)
It runs on my device. What is the question? Are you getting an error?

-- 

David Winsemius, MD
West Hartford, CT