[R] bootstrap

arun smartpink111 at yahoo.com
Wed May 29 16:47:34 CEST 2013



Hi,
Don't know if this helps:

set.seed(244)
 data1<- generate(10)

data1
            x1            x2         x3         x4          x5         x6
1  -0.89558948 -0.5609722069 -1.0782688 -1.2461548 -0.58857050  1.5708187
2  -0.82379187 -0.9056306209 -1.5700755 -0.6045090 -0.19683863 -0.8969354
3  -1.04303377 -0.2379109244 -0.6589381 -0.7521561  1.37492952 -2.7515409
4   0.00783492  0.4249100106 -0.8221292 -0.6220925  0.91199682  0.6365096
5   0.72986110  0.5355575043  1.7584239  1.4474157 -0.46647958 -0.3799255
6  -1.83691611  1.4025663889 -1.0857996  1.3311993 -2.32091759  0.5866616
7   0.10720511  0.6393573943 -0.2811223  2.2800918  0.09360948  0.9441684
8  -0.36297232  0.6974249024  0.9024777  2.0952564  0.22731336  0.7809357
9  -1.28146538  0.0002775273  1.5413933 -3.2237494 -1.31165562  0.6511996
10 -0.65215069  1.8909461951 -0.7355757  0.3532028  0.05926608  1.6003263
            x7         x8         x9        x10 t         y
1  -1.24153055  0.1660665 -0.1791897  0.9132159 0 -1.437598
2   0.76953162  0.6579521 -0.2985585 -0.4716139 0 -2.474706
3  -0.70071814  1.4526669  0.4641041 -1.7046221 1 -0.182020
4   0.70752895 -0.1725368  0.8811518 -0.8023556 0 -3.658734
5  -0.02086093  2.1676000  1.8696969  0.4655824 0 -5.059184
6  -0.39706026  1.4280016  0.6295243 -2.2591846 1 -3.229792
7  -0.47859335  1.9391822  1.1545306 -0.3825239 0 -3.989109
8  -0.96075661  2.4411561 -1.9833116 -0.2589653 0 -1.998027
9  -2.82823504  0.1234223  1.1404231  1.3262285 0 -5.323757
10  0.37175204 -1.4488875 -0.2680561 -2.2540178 0 -5.818188


res1<-as.data.frame(t(sample(as.data.frame(t(data1)),20,replace=TRUE)))
row.names(res1)<-1:nrow(res1)
head(res1,2)
#           x1       x2         x3         x4         x5        x6       x7
#1  0.00783492 0.424910 -0.8221292 -0.6220925 0.91199682 0.6365096 0.707529
#2 -0.65215069 1.890946 -0.7355757  0.3532028 0.05926608 1.6003263 0.371752
 #         x8         x9        x10 t         y
#1 -0.1725368  0.8811518 -0.8023556 0 -3.658734
#2 -1.4488875 -0.2680561 -2.2540178 0 -5.818188
A.K.
________________________________
From: Angela Fel Padecio <solewolf301 at yahoo.com>
To: arun <smartpink111 at yahoo.com> 
Sent: Wednesday, May 29, 2013 10:33 AM
Subject: Re: bootstrap



hi. thanks for the reply. generally, the command for bootstrapping in R is 

sample(x, size, replace=T)

however, i want my x to be the vectors x1 to x10, t and y. i can't do it because i this is what i get:

for instance,
simdata<-sample(data, 20, replace=T)

simdata

                 x6               x7                   x8                   x10                x9               x8.1
1 -1.53816891  1.2408110  0.06074567  0.98045523  1.86121959  0.06074567
2  1.33138543 -1.7287688  0.21719209 -0.03431542  0.69361696  0.21719209
3  0.50121104 -2.1667913  0.29435635  2.75808453 -0.07115435  0.29435635
4 -0.87952116  0.8968212 -1.10899986  0.02983622  0.57830958 -1.10899986
5 -0.03325816 -0.2172203 -1.23341708 -0.71446145  0.91041824 -1.23341708
                      x2              x3      t       x3.1           x1                    x1.1        t.1         x7.1
1  1.971549999 0.7930340 1 0.7930340  1.4209575  1.4209575   1  1.2408110
2 -0.513132396 2.3926484 0 2.3926484 -0.7346689 -0.7346689   0 -1.7287688
3  0.408742759 1.2250794 0 1.2250794 -1.0200933 -1.0200933   0 -2.1667913
4 -0.048902349 1.2637351 1 1.2637351 -2.3791083 -2.3791083   1  0.8968212
5  0.008808758 0.2171343 1 0.2171343  1.8278435  1.8278435   1 -0.2172203
                   x10.1         x5               x5.1              x2.1              x10.2             x3.2
1  0.98045523  0.8283620  0.8283620  1.971549999  0.98045523 0.7930340
2 -0.03431542 -0.3461855 -0.3461855 -0.513132396 -0.03431542 2.3926484
3  2.75808453 -0.4661471 -0.4661471  0.408742759  2.75808453 1.2250794
4  0.02983622  0.4343037  0.4343037 -0.048902349  0.02983622 1.2637351
5 -0.71446145  0.9696091  0.9696091  0.008808758 -0.71446145 0.2171343




________________________________
From: arun <smartpink111 at yahoo.com>
To: Angela Fel Padecio <solewolf301 at yahoo.com> 
Cc: R help <r-help at r-project.org> 
Sent: Wednesday, May 29, 2013 9:15 PM
Subject: Re: bootstrap


Hi,
You might need to check library(boot).  I have never used that before.  So, I can't comment much.  It is better to post on R-help list.  I had seen your postings on Nabble in the past.  Unfortunately those postings were not accepted in R-help.  You have to directly post at   r-help at r-project.org after registering at:
https://stat.ethz.ch/mailman/listinfo/r-help






________________________________
From: Angela Fel Padecio <solewolf301 at yahoo.com>
To: arun <smartpink111 at yahoo.com> 
Sent: Wednesday, May 29, 2013 9:05 AM
Subject: bootstrap



hi. i have these dataset: 

set.seed(12345) 
S=1000 
generate <- function(size) { 
x1 <- rnorm(size, mean=0, sd=1) 
x2 <- rnorm(size, mean=0, sd=1) 
x3 <- rnorm(size, mean=0, sd=1) 
x4 <- rnorm(size, mean=0, sd=1) 
x5 <- rnorm(size, mean=0, sd=1) 
x6 <-
rnorm(size, mean=0, sd=1) 
x7 <- rnorm(size, mean=0, sd=1) 
x8 <- rnorm(size, mean=0, sd=1) 
x9 <- rnorm(size, mean=0, sd=1) 
x10 <- rnorm(size, mean=0, sd=1) 
e<-rnorm(size, mean=0, sd=1) 
t_trueps <- (1 + exp( -(b0 + b1*x1 + b2*x2 + b3*x3 + b4*x4 + b5*x5 + b6*x6 + b7*x7 
+ b2*x2*x2 + b4*x4*x4 + b7*x7*x7 + b1*0.5*x1*x3 + b2*0.7*x2*x4 +b3*0.5*x3*x5 
+ b4*0.7*x4*x6 + b5*0.5*x5*x7 + b1*0.5*x1*x6 + b2*0.7*x2*x3 + b3*0.5*x3*x4 
+ b4*0.5*x4*x5 + b5*0.5*x5*x6) ) )^-1 
prob.exposure <- runif(size) 
t <- ifelse(t_trueps > prob.exposure, 1, 0) 
y <- a0 + a1*x1 + a2*x2 + a3*x3 + a4*x4 +a5*x8 + a6*x9 + a7*x10 + g1*t + e 
sim <- as.data.frame(cbind(x1, x2, x3 ,x4, x5, x6, x7, x8, x9, x10, t, y)) 
return(sim) 
} 
b0 <- 0.05 
b1 <- 0.95 
b2 <- -0.25 
b3 <- 0.6 
b4
<- -0.4 
b5 <- -0.8 
b6 <- -0.5 
b7 <- 0.7 
a0 <- -3.85 
a1 <- 0.3 
a2 <- -0.36 
a3 <- -0.73 
a4 <- -0.2 
a5 <- 0.71 
a6 <- -0.19 
a7 <- 0.26 
g1 <- -0.4 

How can i have a bootstrap sample of 5 from size 10 with 1000 replications, i.e, 

data<-generate(10) 

where entire row is chosen. for example, if i have this result: 

> data<-generate(10) 
> data 
           x1          x2         x3          x4         x5         x6 
1   0.9374775 -2.26118020 -1.7502526 -0.04386488  0.6760577  0.5125265 
2  -1.3379878  0.54495810 -1.1884175 -0.25142873 -0.7389659
-2.7850113 
3  -0.5633375 -0.59191774 -1.4772120  0.74107022 -0.2862497  0.8067138 
4   0.3050390 -0.40174654  0.2980965  0.19234445  1.0762513  0.1604186 
5   0.6919353 -0.13762332 -0.5352159  0.69566392  1.3655018 -0.8463535 
6   0.2988476  1.15431908 -0.1172911  1.38397105  0.2631057  1.7943873 
7  -0.1424191 -0.06971174 -1.2759867 -1.25774074  1.2320712  1.7419884 
8   1.1835778 -0.06363542 -0.4772019 -0.46848004  1.0064428 -0.6327513 
9   0.1942304 -0.79500297 -1.1120049  0.60023523  0.2210129 -1.7187271 
10 -2.3568380  0.34227984 -1.0006803 -0.09190608  0.1187722 -0.5756797 
           x7                  x8                x9  
              x10           t         y 
1   0.7249311  0.2420102  2.51106406  1.6699921 0 -2.019059 
2  -0.5209441  0.4935232 -0.31236000 -0.7717535 1 -2.286814 
3  -1.0086591  0.6120889 -0.19684965  0.3355901 0 -2.785737 
4  -0.4003396  0.9890145 -2.10427706  0.5364553 1 -2.782166 
5  -0.1662400 -0.1357740  0.80220396 -0.3509522 0 -2.424431 
6   2.1162941  1.4961478  0.02505993  0.8154658 1 -6.140725 
7   0.9965705  1.0531386  0.53474097 -0.1415918 0 -2.922637 
8  -0.1959067 -0.3507144  1.15341204  0.6018451 1 -2.602453 
9   1.0153402 -0.2612589 -0.70037063  0.4304613 1 -2.788079 
10  0.5378466  0.8504346  1.66384979 -1.2695204 0
-3.749164 

entire entries of row 1,3,5,7,9 for instance are chosen with 1000. your help is highly appreciated.  



More information about the R-help mailing list