R help - Nov 2011 - Listing tables together from random samples from a generated population?

.
HI there,
I'd like to show demonstrate how the chi-squared distribution works, so
I've come up with a sample data frame of two categorical variables
y<-data.frame(gender=sample(c('Male', 'Female'), size=100000,
replace=TRUE, c(0.5, 0.5)), tea=sample(c('Yes', 'No'),
size=100000, replace=TRUE, c(0.5, 0.5)))

And I'd like to create a list of 100 different samples of those two
variables and the resulting 2X2 contingency tables

table(.y[sample(nrow(.y), 100), ])

How would I combine these 100 tables into a list? I'd like to be able to go
in and find some of the extreme values to show how the sampling distribution of
the chi-square values.

I can already get a histogram of 100 different chi-squared values that shows the
distribution nicely (see below), but I'd like to actually show the
underlying tables, for demonstration's sake.

 .z<-vector()
for (i in 1:100) {
.z<-c(.z, chisq.test(table(.y[sample(nrow(.y), 200), ]))$statistic)
}
hist(.z, xlab='Chi-Square Value', main="Chi-Squared Values From 100
different samples asking\nabout gender and tea/coffee drinking")
abline(v=3.84, lty=2)

Thank you in advance,
Simon Kiss

*********************************
Simon J. Kiss, PhD
Assistant Professor, Wilfrid Laurier University
73 George Street
Brantford, Ontario, Canada
N3T 2C9
Cell: +1 905 746 7606

You could try something like this:
        tables <- lapply(seq(100), function(i) table(.y[sample(nrow(.y), 
200), ]))

Then you could conduct the chi-squared tests
        chisqs <- lapply(tables, chisq.test)
and save the values
        .z <- sapply(chisqs, "[[", "statistic")

Jean

---

Simon Kiss wrote on 11/10/2011 15:48:38:

HI there, 
I'd like to show demonstrate how the chi-squared distribution works, so 
I've come up with a sample data frame of two categorical variables 
y<-data.frame(gender=sample(c('Male', 'Female'), size=100000,
replace=TRUE, c(0.5, 0.5)), tea=sample(c('Yes', 'No'),
size=100000,
replace=TRUE, c(0.5, 0.5)))

And I'd like to create a list of 100 different samples of those two 
variables and the resulting 2X2 contingency tables

table(.y[sample(nrow(.y), 100), ])

How would I combine these 100 tables into a list? I'd like to be able to 
go in and find some of the extreme values to show how the sampling 
distribution of the chi-square values.

I can already get a histogram of 100 different chi-squared values that 
shows the distribution nicely (see below), but I'd like to actually show 
the underlying tables, for demonstration's sake.

 .z<-vector() 
for (i in 1:100) { 
.z<-c(.z, chisq.test(table(.y[sample(nrow(.y), 200), ]))$statistic) 
} 
hist(.z, xlab='Chi-Square Value', main="Chi-Squared Values From 100
different samples asking\nabout gender and tea/coffee drinking") 
abline(v=3.84, lty=2)

Thank you in advance, 
Simon Kiss

Simon J. Kiss, PhD 
Assistant Professor, Wilfrid Laurier University 73 George Street 
Brantford, Ontario, Canada 
N3T 2C9 
Cell: +1 905 746 7606
	[[alternative HTML version deleted]]

Hi:

There are two ways you could go about this: lists or arrays. It's
pretty easy to generate an array, a little more work to get the list.
I'm assuming the objective is to extract a chi-square statistic from
each table, so I'll show a couple of ways to do that, too.

library('plyr')

## Start with the data:
y<-data.frame(gender=sample(c('Male', 'Female'), size=100000,
                                            replace=TRUE, c(0.5, 0.5)),
                     tea=sample(c('Yes', 'No'),  size=100000,
                                        replace=TRUE, c(0.5, 0.5)))
## Function to produce a table:
tabfun <- function(d) table(d[sample(seq_len(nrow(d)), 100), ])
x2stat <- function(m) chisq.test(m)$statistic

## Array version:

tbarr <- replicate(100, tabfun(y))
# X^2 statistics using apply() from base R and
# aaply() from plyr:
u1 <- apply(tablist, 3, x2stat)
u2 <- aaply(tablist, 3, x2stat)

## List version:

tblst <- vector('list', 100)
for(i in seq_along(tblst)) tblst[[i]] <- tabfun(y)

v1 <- unname(do.call(c, lapply(tblst, x2stat)))
v2 <- laply(tblst, x2stat)
>From here, it's easy to do the histogram :)
HTH,
Dennis


On Thu, Nov 10, 2011 at 12:48 PM, Simon Kiss <sjkiss at gmail.com>
wrote:
> .
> HI there,
> I'd like to show demonstrate how the chi-squared distribution works, so
I've come up with a sample data frame of two categorical variables
> y<-data.frame(gender=sample(c('Male', 'Female'),
size=100000, replace=TRUE, c(0.5, 0.5)), tea=sample(c('Yes',
'No'), size=100000, replace=TRUE, c(0.5, 0.5)))
>
> And I'd like to create a list of 100 different samples of those two
variables and the resulting 2X2 contingency tables
>
> table(.y[sample(nrow(.y), 100), ])
>
> How would I combine these 100 tables into a list? I'd like to be able
to go in and find some of the extreme values to show how the sampling
distribution of the chi-square values.
>
> I can already get a histogram of 100 different chi-squared values that
shows the distribution nicely (see below), but I'd like to actually show the
underlying tables, for demonstration's sake.
>
> ?.z<-vector()
> for (i in 1:100) {
> .z<-c(.z, chisq.test(table(.y[sample(nrow(.y), 200), ]))$statistic)
> }
> hist(.z, xlab='Chi-Square Value', main="Chi-Squared Values
From 100 different samples asking\nabout gender and tea/coffee drinking")
> abline(v=3.84, lty=2)
>
> Thank you in advance,
> Simon Kiss
>
> *********************************
> Simon J. Kiss, PhD
> Assistant Professor, Wilfrid Laurier University
> 73 George Street
> Brantford, Ontario, Canada
> N3T 2C9
> Cell: +1 905 746 7606
>
> ______________________________________________
> 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.
>