R help - Jul 2004 - extracting the t-statistic: just the numbers, please

Hi, there

I am quite sure there is an easy answer to this, but I am unsure how to gather 
a bunch of t-statistics in an organized format.

I am trying to generate a list of t-statistics for a randomization routine. If 
I try to collect a bunch of t-statistics from a run, this is what happens:

> M <- 10 ; simt <- NULL
> for(i in 1:M)
+     {
+    perm<-sample(site,replace=F)
+ 
+ permute<-cbind(perm, site, a, b, c)
+ 
+ m<- order(perm)
+ 
+ m1<-cbind(perm[m], site[m], a[m], b[m], c[m])
+ 
+ black<-c((m1[1:5,5]),(m1[11:15,5]))
+ #black
+ 
+ white<-c((m1[6:10,5]),(m1[16:20,5]))
+ #white
+ 
+ sims <- t.test(black,white,var.equal=FALSE,mu=0)$statistic
+ simt<-c(simt,sims)
+ #simt
+ } # Next i (next simulation)
> 
> simt
         t          t          t          t          t          t          t 
 0.3474150  0.1542973 -0.4044992  1.2466663 -0.2933944 -0.5809257  0.7799080 
         t          t          t 
-1.4132713  1.2048335 -0.6596936

Which gives me a list, but not in a form that I can do anything with. This is 
in stark contrast to what happens when requesting p-values, which gives output 
like this:



 M <- 10 ; simt <- NULL
> for(i in 1:M)
+     {
+    perm<-sample(site,replace=F)
+ 
+ permute<-cbind(perm, site, a, b, c)
+ 
+ m<- order(perm)
+ 
+ m1<-cbind(perm[m], site[m], a[m], b[m], c[m])
+ 
+ black<-c((m1[1:5,5]),(m1[11:15,5]))
+ #black
+ 
+ white<-c((m1[6:10,29]),(m1[16:20,5]))
+ #white
+ 
+ sims <- t.test(black,white,var.equal=FALSE,mu=0)$p.value
+ simt<-c(simt,sims)
+ #simt
+ } # Next i (next simulation)
> 
> simt
 [1] 0.6763749 0.7480091 0.9447851 0.3342029 0.7852635 0.3199006 0.5272153
 [8] 0.3863616 0.7333693 0.7268907

Now THAT'S what I'd like to get for my t-statistics- a nice vector
(simt) that
I can deal with later, rather than the output I am currently getting (the first 
output above).

Does anyone know a way to extract JUST the t-statistics from the t.test, 
without the "t" character header, so I can generate a nice little
vector?
Alternatively, can I manipulate the output I am currently getting for the t-
statistics so that I can isolate just the numbers? 

-- 
Michael Rennie
Ph.D. Candidate
University of Toronto at Mississauga
3359 Mississauga Rd. N.
Mississauga ON  L5L 1C6
Ph: 905-828-5452  Fax: 905-828-3792

>	data(sleep)
>      t.test(extra ~ group, data = sleep)$statistic[1]
        t
-1.860813
>      t.test(extra ~ group, data = sleep)$statistic[[1]]
[1] -1.860813

hope this helps,
christian


Am Donnerstag, 29. Juli 2004 07:57 schrieb Michael
Rennie:
> Hi, there
>
> I am quite sure there is an easy answer to this, but I am unsure how to
> gather a bunch of t-statistics in an organized format.
>
> I am trying to generate a list of t-statistics for a randomization routine.
> If
>
> I try to collect a bunch of t-statistics from a run, this is what happens:
> > M <- 10 ; simt <- NULL
> > for(i in 1:M)
>
> +     {
> +    perm<-sample(site,replace=F)
> +
> + permute<-cbind(perm, site, a, b, c)
> +
> + m<- order(perm)
> +
> + m1<-cbind(perm[m], site[m], a[m], b[m], c[m])
> +
> + black<-c((m1[1:5,5]),(m1[11:15,5]))
> + #black
> +
> + white<-c((m1[6:10,5]),(m1[16:20,5]))
> + #white
> +
> + sims <- t.test(black,white,var.equal=FALSE,mu=0)$statistic
> + simt<-c(simt,sims)
> + #simt
> + } # Next i (next simulation)
>
> > simt
>
>          t          t          t          t          t          t         
> t 0.3474150  0.1542973 -0.4044992  1.2466663 -0.2933944 -0.5809257 
> 0.7799080 t          t          t
> -1.4132713  1.2048335 -0.6596936
>
> Which gives me a list, but not in a form that I can do anything with. This
> is in stark contrast to what happens when requesting p-values, which gives
> output like this:
>
>
>
>  M <- 10 ; simt <- NULL
>
> > for(i in 1:M)
>
> +     {
> +    perm<-sample(site,replace=F)
> +
> + permute<-cbind(perm, site, a, b, c)
> +
> + m<- order(perm)
> +
> + m1<-cbind(perm[m], site[m], a[m], b[m], c[m])
> +
> + black<-c((m1[1:5,5]),(m1[11:15,5]))
> + #black
> +
> + white<-c((m1[6:10,29]),(m1[16:20,5]))
> + #white
> +
> + sims <- t.test(black,white,var.equal=FALSE,mu=0)$p.value
> + simt<-c(simt,sims)
> + #simt
> + } # Next i (next simulation)
>
> > simt
>
>  [1] 0.6763749 0.7480091 0.9447851 0.3342029 0.7852635 0.3199006 0.5272153
>  [8] 0.3863616 0.7333693 0.7268907
>
> Now THAT'S what I'd like to get for my t-statistics- a nice vector
(simt)
> that I can deal with later, rather than the output I am currently getting
> (the first output above).
>
> Does anyone know a way to extract JUST the t-statistics from the t.test,
> without the "t" character header, so I can generate a nice little
vector?
> Alternatively, can I manipulate the output I am currently getting for the
> t- statistics so that I can isolate just the numbers?

The function "as.vector" will remove the names: 

 > t.test(1:9)$statistic
       t
5.477226
 > as.vector(t.test(1:9)$statistic)
[1] 5.477226

      Is this what you want?  spencer graves

Michael Rennie wrote:
>Hi, there
>
>I am quite sure there is an easy answer to this, but I am unsure how to
gather
>a bunch of t-statistics in an organized format.
>
>I am trying to generate a list of t-statistics for a randomization routine.
If
>I try to collect a bunch of t-statistics from a run, this is what happens:
>
>
>  
>
>>M <- 10 ; simt <- NULL
>>for(i in 1:M)
>>    
>>
>+     {
>+    perm<-sample(site,replace=F)
>+ 
>+ permute<-cbind(perm, site, a, b, c)
>+ 
>+ m<- order(perm)
>+ 
>+ m1<-cbind(perm[m], site[m], a[m], b[m], c[m])
>+ 
>+ black<-c((m1[1:5,5]),(m1[11:15,5]))
>+ #black
>+ 
>+ white<-c((m1[6:10,5]),(m1[16:20,5]))
>+ #white
>+ 
>+ sims <- t.test(black,white,var.equal=FALSE,mu=0)$statistic
>+ simt<-c(simt,sims)
>+ #simt
>+ } # Next i (next simulation)
>  
>
>>simt
>>    
>>
>         t          t          t          t          t          t          t
> 0.3474150  0.1542973 -0.4044992  1.2466663 -0.2933944 -0.5809257  0.7799080
>         t          t          t 
>-1.4132713  1.2048335 -0.6596936
>
>Which gives me a list, but not in a form that I can do anything with. This
is
>in stark contrast to what happens when requesting p-values, which gives
output
>like this:
>
>
>
> M <- 10 ; simt <- NULL
>  
>
>>for(i in 1:M)
>>    
>>
>+     {
>+    perm<-sample(site,replace=F)
>+ 
>+ permute<-cbind(perm, site, a, b, c)
>+ 
>+ m<- order(perm)
>+ 
>+ m1<-cbind(perm[m], site[m], a[m], b[m], c[m])
>+ 
>+ black<-c((m1[1:5,5]),(m1[11:15,5]))
>+ #black
>+ 
>+ white<-c((m1[6:10,29]),(m1[16:20,5]))
>+ #white
>+ 
>+ sims <- t.test(black,white,var.equal=FALSE,mu=0)$p.value
>+ simt<-c(simt,sims)
>+ #simt
>+ } # Next i (next simulation)
>  
>
>>simt
>>    
>>
> [1] 0.6763749 0.7480091 0.9447851 0.3342029 0.7852635 0.3199006 0.5272153
> [8] 0.3863616 0.7333693 0.7268907
>
>Now THAT'S what I'd like to get for my t-statistics- a nice vector
(simt) that
>I can deal with later, rather than the output I am currently getting (the
first
>output above).
>
>Does anyone know a way to extract JUST the t-statistics from the t.test, 
>without the "t" character header, so I can generate a nice little
vector?
>Alternatively, can I manipulate the output I am currently getting for the t-
>statistics so that I can isolate just the numbers? 
>
>  
>

Michael Rennie <mrennie at utm.utoronto.ca> writes:
> > simt
>          t          t          t          t          t          t         
t
>  0.3474150  0.1542973 -0.4044992  1.2466663 -0.2933944 -0.5809257 
0.7799080
>          t          t          t 
> -1.4132713  1.2048335 -0.6596936
> 
> Which gives me a list, but not in a form that I can do anything with. This
is
> in stark contrast to what happens when requesting p-values, which gives
output
> like this:
..
> > simt
>  [1] 0.6763749 0.7480091 0.9447851 0.3342029 0.7852635 0.3199006 0.5272153
>  [8] 0.3863616 0.7333693 0.7268907
> 
> Now THAT'S what I'd like to get for my t-statistics- a nice vector
(simt) that
> I can deal with later, rather than the output I am currently getting (the
first
> output above).
> 
> Does anyone know a way to extract JUST the t-statistics from the t.test, 
> without the "t" character header, so I can generate a nice little
vector?
> Alternatively, can I manipulate the output I am currently getting for the
t-
> statistics so that I can isolate just the numbers? 
It *is* a nice little vector! It's just that it has names. 

names(simt) <- NULL 

gets rid of them. 

BTW, simt <- c(simt, sims) is going to kill performance with large repeat
counts since it copies all previous results every time around.
Preallocate the result, or consider using replicate() or sapply().

-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907