Calculate summary statistics grouped by variable
dby(
data,
INPUT,
...,
ID = NULL,
ORDER = NULL,
SUBSET = NULL,
SORT = 0,
COMBINE = !REDUCE,
NOCHECK = FALSE,
ARGS = NULL,
NAMES,
COLUMN = FALSE,
REDUCE = FALSE,
REGEX = mets.options()$regex,
ALL = TRUE
)Data.frame
Input variables (character or formula)
functions
id variable
(optional) order variable
(optional) subset expression
sort order (id+order variable)
If TRUE result is appended to data
No sorting or check for missing data
Optional list of arguments to functions (...)
Optional vector of column names
If TRUE do the calculations for each column
Reduce number of redundant rows
Allow regular expressions
if FALSE only the subset will be returned
Calculate summary statistics grouped by
dby2 for column-wise calculations
n <- 4
k <- c(3,rbinom(n-1,3,0.5)+1)
N <- sum(k)
d <- data.frame(y=rnorm(N),x=rnorm(N),
id=rep(seq(n),k),num=unlist(sapply(k,seq))
)
d2 <- d[sample(nrow(d)),]
dby(d2, y~id, mean)
#> y x id num mean
#> 1 -0.65666950 0.3064051 1 1 0.5684420
#> 3 2.24915454 -1.1170657 1 3 0.5684420
#> 2 0.11284083 0.0211384 1 2 0.5684420
#> 6 0.58410126 -0.1862686 2 3 -0.2496999
#> 5 -0.35545945 -0.7846569 2 2 -0.2496999
#> 4 -0.97774138 -1.5214842 2 1 -0.2496999
#> 7 0.05404795 -0.3898105 3 1 -0.5529647
#> 8 -1.15997739 -1.1005508 3 2 -0.5529647
#> 10 -1.12149743 -0.6417276 4 2 -0.6258272
#> 11 -0.18221248 0.9872356 4 3 -0.6258272
#> 9 -0.57377160 0.5216237 4 1 -0.6258272
dby(d2, y~id + order(num), cumsum)
#> y x id num cumsum
#> 1 -0.65666950 0.3064051 1 1 -0.65666950
#> 2 0.11284083 0.0211384 1 2 -0.54382867
#> 3 2.24915454 -1.1170657 1 3 1.70532587
#> 4 -0.97774138 -1.5214842 2 1 -0.97774138
#> 5 -0.35545945 -0.7846569 2 2 -1.33320083
#> 6 0.58410126 -0.1862686 2 3 -0.74909958
#> 7 0.05404795 -0.3898105 3 1 0.05404795
#> 8 -1.15997739 -1.1005508 3 2 -1.10592944
#> 9 -0.57377160 0.5216237 4 1 -0.57377160
#> 10 -1.12149743 -0.6417276 4 2 -1.69526903
#> 11 -0.18221248 0.9872356 4 3 -1.87748151
dby(d,y ~ id + order(num), dlag)
#> y x id num dlag
#> 1 -0.65666950 0.3064051 1 1 NA
#> 2 0.11284083 0.0211384 1 2 -0.65666950
#> 3 2.24915454 -1.1170657 1 3 0.11284083
#> 4 -0.97774138 -1.5214842 2 1 NA
#> 5 -0.35545945 -0.7846569 2 2 -0.97774138
#> 6 0.58410126 -0.1862686 2 3 -0.35545945
#> 7 0.05404795 -0.3898105 3 1 NA
#> 8 -1.15997739 -1.1005508 3 2 0.05404795
#> 9 -0.57377160 0.5216237 4 1 NA
#> 10 -1.12149743 -0.6417276 4 2 -0.57377160
#> 11 -0.18221248 0.9872356 4 3 -1.12149743
dby(d,y ~ id + order(num), dlag, ARGS=list(k=1:2))
#> y x id num dlag1 dlag2
#> 1 -0.65666950 0.3064051 1 1 NA NA
#> 2 0.11284083 0.0211384 1 2 -0.65666950 NA
#> 3 2.24915454 -1.1170657 1 3 0.11284083 -0.6566695
#> 4 -0.97774138 -1.5214842 2 1 NA NA
#> 5 -0.35545945 -0.7846569 2 2 -0.97774138 NA
#> 6 0.58410126 -0.1862686 2 3 -0.35545945 -0.9777414
#> 7 0.05404795 -0.3898105 3 1 NA NA
#> 8 -1.15997739 -1.1005508 3 2 0.05404795 NA
#> 9 -0.57377160 0.5216237 4 1 NA NA
#> 10 -1.12149743 -0.6417276 4 2 -0.57377160 NA
#> 11 -0.18221248 0.9872356 4 3 -1.12149743 -0.5737716
dby(d,y ~ id + order(num), dlag, ARGS=list(k=1:2), NAMES=c("l1","l2"))
#> y x id num l1 l2
#> 1 -0.65666950 0.3064051 1 1 NA NA
#> 2 0.11284083 0.0211384 1 2 -0.65666950 NA
#> 3 2.24915454 -1.1170657 1 3 0.11284083 -0.6566695
#> 4 -0.97774138 -1.5214842 2 1 NA NA
#> 5 -0.35545945 -0.7846569 2 2 -0.97774138 NA
#> 6 0.58410126 -0.1862686 2 3 -0.35545945 -0.9777414
#> 7 0.05404795 -0.3898105 3 1 NA NA
#> 8 -1.15997739 -1.1005508 3 2 0.05404795 NA
#> 9 -0.57377160 0.5216237 4 1 NA NA
#> 10 -1.12149743 -0.6417276 4 2 -0.57377160 NA
#> 11 -0.18221248 0.9872356 4 3 -1.12149743 -0.5737716
dby(d, y~id + order(num), mean=mean, csum=cumsum, n=length)
#> y x id num mean csum n
#> 1 -0.65666950 0.3064051 1 1 0.5684420 -0.65666950 3
#> 2 0.11284083 0.0211384 1 2 0.5684420 -0.54382867 3
#> 3 2.24915454 -1.1170657 1 3 0.5684420 1.70532587 3
#> 4 -0.97774138 -1.5214842 2 1 -0.2496999 -0.97774138 3
#> 5 -0.35545945 -0.7846569 2 2 -0.2496999 -1.33320083 3
#> 6 0.58410126 -0.1862686 2 3 -0.2496999 -0.74909958 3
#> 7 0.05404795 -0.3898105 3 1 -0.5529647 0.05404795 2
#> 8 -1.15997739 -1.1005508 3 2 -0.5529647 -1.10592944 2
#> 9 -0.57377160 0.5216237 4 1 -0.6258272 -0.57377160 3
#> 10 -1.12149743 -0.6417276 4 2 -0.6258272 -1.69526903 3
#> 11 -0.18221248 0.9872356 4 3 -0.6258272 -1.87748151 3
dby(d2, y~id + order(num), a=cumsum, b=mean, N=length,
l1=function(x) c(NA,x)[-length(x)]
)
#> y x id num a b N l1
#> 1 -0.65666950 0.3064051 1 1 -0.65666950 0.5684420 3 NA
#> 2 0.11284083 0.0211384 1 2 -0.54382867 0.5684420 3 -0.6566695
#> 3 2.24915454 -1.1170657 1 3 1.70532587 0.5684420 3 2.2491545
#> 4 -0.97774138 -1.5214842 2 1 -0.97774138 -0.2496999 3 NA
#> 5 -0.35545945 -0.7846569 2 2 -1.33320083 -0.2496999 3 -0.9777414
#> 6 0.58410126 -0.1862686 2 3 -0.74909958 -0.2496999 3 0.5841013
#> 7 0.05404795 -0.3898105 3 1 0.05404795 -0.5529647 2 NA
#> 8 -1.15997739 -1.1005508 3 2 -1.10592944 -0.5529647 2 -1.1599774
#> 9 -0.57377160 0.5216237 4 1 -0.57377160 -0.6258272 3 NA
#> 10 -1.12149743 -0.6417276 4 2 -1.69526903 -0.6258272 3 -0.5737716
#> 11 -0.18221248 0.9872356 4 3 -1.87748151 -0.6258272 3 -0.1822125
dby(d, y~id + order(num), nn=seq_along, n=length)
#> y x id num nn n
#> 1 -0.65666950 0.3064051 1 1 1 3
#> 2 0.11284083 0.0211384 1 2 2 3
#> 3 2.24915454 -1.1170657 1 3 3 3
#> 4 -0.97774138 -1.5214842 2 1 1 3
#> 5 -0.35545945 -0.7846569 2 2 2 3
#> 6 0.58410126 -0.1862686 2 3 3 3
#> 7 0.05404795 -0.3898105 3 1 1 2
#> 8 -1.15997739 -1.1005508 3 2 2 2
#> 9 -0.57377160 0.5216237 4 1 1 3
#> 10 -1.12149743 -0.6417276 4 2 2 3
#> 11 -0.18221248 0.9872356 4 3 3 3
dby(d, y~id + order(num), nn=seq_along, n=length)
#> y x id num nn n
#> 1 -0.65666950 0.3064051 1 1 1 3
#> 2 0.11284083 0.0211384 1 2 2 3
#> 3 2.24915454 -1.1170657 1 3 3 3
#> 4 -0.97774138 -1.5214842 2 1 1 3
#> 5 -0.35545945 -0.7846569 2 2 2 3
#> 6 0.58410126 -0.1862686 2 3 3 3
#> 7 0.05404795 -0.3898105 3 1 1 2
#> 8 -1.15997739 -1.1005508 3 2 2 2
#> 9 -0.57377160 0.5216237 4 1 1 3
#> 10 -1.12149743 -0.6417276 4 2 2 3
#> 11 -0.18221248 0.9872356 4 3 3 3
d <- d[,1:4]
dby(d, x<0) <- list(z=mean)
d <- dby(d, is.na(z), z=1)
f <- function(x) apply(x,1,min)
dby(d, y+x~id, min=f)
#> Error: object 'f' not found
dby(d,y+x~id+order(num), function(x) x)
#> y x id num z _11 _12
#> 1 -0.65666950 0.3064051 1 1 1.0000000 -0.65666950 0.3064051
#> 2 0.11284083 0.0211384 1 2 1.0000000 0.11284083 0.0211384
#> 3 2.24915454 -1.1170657 1 3 0.8761094 2.24915454 -1.1170657
#> 4 -0.97774138 -1.5214842 2 1 0.8761094 -0.97774138 -1.5214842
#> 5 -0.35545945 -0.7846569 2 2 0.8761094 -0.35545945 -0.7846569
#> 6 0.58410126 -0.1862686 2 3 0.8761094 0.58410126 -0.1862686
#> 7 0.05404795 -0.3898105 3 1 0.8761094 0.05404795 -0.3898105
#> 8 -1.15997739 -1.1005508 3 2 0.8761094 -1.15997739 -1.1005508
#> 9 -0.57377160 0.5216237 4 1 1.0000000 -0.57377160 0.5216237
#> 10 -1.12149743 -0.6417276 4 2 0.8761094 -1.12149743 -0.6417276
#> 11 -0.18221248 0.9872356 4 3 1.0000000 -0.18221248 0.9872356
f <- function(x) { cbind(cumsum(x[,1]),cumsum(x[,2]))/sum(x)}
dby(d, y+x~id, f)
#> Error: object 'f' not found
## column-wise
a <- d
dby2(a, mean, median, REGEX=TRUE) <- '^[y|x]'~id
a
#> y x id num z mean.y mean.x median.y
#> 1 -0.65666950 0.3064051 1 1 1.0000000 0.5684420 -0.2631740 0.1128408
#> 2 0.11284083 0.0211384 1 2 1.0000000 0.5684420 -0.2631740 0.1128408
#> 3 2.24915454 -1.1170657 1 3 0.8761094 0.5684420 -0.2631740 0.1128408
#> 4 -0.97774138 -1.5214842 2 1 0.8761094 -0.2496999 -0.8308032 -0.3554594
#> 5 -0.35545945 -0.7846569 2 2 0.8761094 -0.2496999 -0.8308032 -0.3554594
#> 6 0.58410126 -0.1862686 2 3 0.8761094 -0.2496999 -0.8308032 -0.3554594
#> 7 0.05404795 -0.3898105 3 1 0.8761094 -0.5529647 -0.7451807 -0.5529647
#> 8 -1.15997739 -1.1005508 3 2 0.8761094 -0.5529647 -0.7451807 -0.5529647
#> 9 -0.57377160 0.5216237 4 1 1.0000000 -0.6258272 0.2890439 -0.5737716
#> 10 -1.12149743 -0.6417276 4 2 0.8761094 -0.6258272 0.2890439 -0.5737716
#> 11 -0.18221248 0.9872356 4 3 1.0000000 -0.6258272 0.2890439 -0.5737716
#> median.x
#> 1 0.0211384
#> 2 0.0211384
#> 3 0.0211384
#> 4 -0.7846569
#> 5 -0.7846569
#> 6 -0.7846569
#> 7 -0.7451807
#> 8 -0.7451807
#> 9 0.5216237
#> 10 0.5216237
#> 11 0.5216237
## wildcards
dby2(a,'y*'+'x*'~id,mean)
#> y x id num z median.y median.x mean.y
#> 1 -0.65666950 0.3064051 1 1 1.0000000 0.1128408 0.0211384 0.5684420
#> 2 0.11284083 0.0211384 1 2 1.0000000 0.1128408 0.0211384 0.5684420
#> 3 2.24915454 -1.1170657 1 3 0.8761094 0.1128408 0.0211384 0.5684420
#> 4 -0.97774138 -1.5214842 2 1 0.8761094 -0.3554594 -0.7846569 -0.2496999
#> 5 -0.35545945 -0.7846569 2 2 0.8761094 -0.3554594 -0.7846569 -0.2496999
#> 6 0.58410126 -0.1862686 2 3 0.8761094 -0.3554594 -0.7846569 -0.2496999
#> 7 0.05404795 -0.3898105 3 1 0.8761094 -0.5529647 -0.7451807 -0.5529647
#> 8 -1.15997739 -1.1005508 3 2 0.8761094 -0.5529647 -0.7451807 -0.5529647
#> 9 -0.57377160 0.5216237 4 1 1.0000000 -0.5737716 0.5216237 -0.6258272
#> 10 -1.12149743 -0.6417276 4 2 0.8761094 -0.5737716 0.5216237 -0.6258272
#> 11 -0.18221248 0.9872356 4 3 1.0000000 -0.5737716 0.5216237 -0.6258272
#> mean.x
#> 1 -0.2631740
#> 2 -0.2631740
#> 3 -0.2631740
#> 4 -0.8308032
#> 5 -0.8308032
#> 6 -0.8308032
#> 7 -0.7451807
#> 8 -0.7451807
#> 9 0.2890439
#> 10 0.2890439
#> 11 0.2890439
## subset
dby(d, x<0) <- list(z=NA)
d
#> y x id num z
#> 1 -0.65666950 0.3064051 1 1 1
#> 2 0.11284083 0.0211384 1 2 1
#> 3 2.24915454 -1.1170657 1 3 NA
#> 4 -0.97774138 -1.5214842 2 1 NA
#> 5 -0.35545945 -0.7846569 2 2 NA
#> 6 0.58410126 -0.1862686 2 3 NA
#> 7 0.05404795 -0.3898105 3 1 NA
#> 8 -1.15997739 -1.1005508 3 2 NA
#> 9 -0.57377160 0.5216237 4 1 1
#> 10 -1.12149743 -0.6417276 4 2 NA
#> 11 -0.18221248 0.9872356 4 3 1
dby(d, y~id|x>-1, v=mean,z=1)
#> y x id num v z
#> 1 -0.65666950 0.3064051 1 1 -0.27191433 1
#> 2 0.11284083 0.0211384 1 2 -0.27191433 1
#> 3 2.24915454 -1.1170657 1 3 NA NA
#> 4 -0.97774138 -1.5214842 2 1 NA NA
#> 5 -0.35545945 -0.7846569 2 2 0.11432090 1
#> 6 0.58410126 -0.1862686 2 3 0.11432090 1
#> 7 0.05404795 -0.3898105 3 1 0.05404795 1
#> 8 -1.15997739 -1.1005508 3 2 NA NA
#> 9 -0.57377160 0.5216237 4 1 -0.62582717 1
#> 10 -1.12149743 -0.6417276 4 2 -0.62582717 1
#> 11 -0.18221248 0.9872356 4 3 -0.62582717 1
dby(d, y+x~id|x>-1, mean, median, COLUMN=TRUE)
#> y x id num z mean.y mean.x median.y
#> 1 -0.65666950 0.3064051 1 1 1 -0.27191433 0.1637718 -0.27191433
#> 2 0.11284083 0.0211384 1 2 1 -0.27191433 0.1637718 -0.27191433
#> 3 2.24915454 -1.1170657 1 3 NA NA NA NA
#> 4 -0.97774138 -1.5214842 2 1 NA NA NA NA
#> 5 -0.35545945 -0.7846569 2 2 NA 0.11432090 -0.4854627 0.11432090
#> 6 0.58410126 -0.1862686 2 3 NA 0.11432090 -0.4854627 0.11432090
#> 7 0.05404795 -0.3898105 3 1 NA 0.05404795 -0.3898105 0.05404795
#> 8 -1.15997739 -1.1005508 3 2 NA NA NA NA
#> 9 -0.57377160 0.5216237 4 1 1 -0.62582717 0.2890439 -0.57377160
#> 10 -1.12149743 -0.6417276 4 2 NA -0.62582717 0.2890439 -0.57377160
#> 11 -0.18221248 0.9872356 4 3 1 -0.62582717 0.2890439 -0.57377160
#> median.x
#> 1 0.1637718
#> 2 0.1637718
#> 3 NA
#> 4 NA
#> 5 -0.4854627
#> 6 -0.4854627
#> 7 -0.3898105
#> 8 NA
#> 9 0.5216237
#> 10 0.5216237
#> 11 0.5216237
dby2(d, y+x~id|x>0, mean, REDUCE=TRUE)
#> id mean.y mean.x
#> 1 1 -0.2719143 0.1637718
#> 2 4 -0.3779920 0.7544297
dby(d,y~id|x<0,mean,ALL=FALSE)
#> y x id num z mean
#> 3 2.24915454 -1.1170657 1 3 NA 2.2491545
#> 4 -0.97774138 -1.5214842 2 1 NA -0.2496999
#> 5 -0.35545945 -0.7846569 2 2 NA -0.2496999
#> 6 0.58410126 -0.1862686 2 3 NA -0.2496999
#> 7 0.05404795 -0.3898105 3 1 NA -0.5529647
#> 8 -1.15997739 -1.1005508 3 2 NA -0.5529647
#> 10 -1.12149743 -0.6417276 4 2 NA -1.1214974
a <- iris
a <- dby(a,y=1)
dby(a,Species=="versicolor") <- list(y=2)