rvsummary
is a class of objects that hold the summary information on
each scalar component of a random variable (quantiles, mean, sd, number of
simulations etc.)
object to be coerced or tested
further arguments passed to or from other methods.
quantiles to calculate and store in the object
logical; whether to print all levels or not (see below for details)
An object of class rvsummary
and of subclass
rvsummary_numeric
, rvsummary_integer
,
rvsummary_logical
, or rvsummary_rvfactor
.
The rvsummary
class provides a means to store a concise
representation of the marginal posterior distributions of the vector
components. By default, the 201 quantiles
0, 0.005, 0.01,
0.015, ..., 0.990, 0.995, 1
are saved for each vector component in an
rvsummary
object.
is.rvsummary
tests whether the object is an rvsummary
object;
as.rvsummary
coerces a random vector object to a rvsummary
object.
as.data.frame
is another way to obtain the data frame that is
produced by the summary
method.
A data frame that has the format of an rv
summary can be coerced into
an rvsummary
; if quantiles are not specified within the data frame,
quantiles from the Normal distribution are filled in, if the mean and s.d.
are given.
Therefore, the following (generic) functions work with rvsummary
objects: rvmean
, rvsd
, rvvar
, rvquantile
,
rnsims
, sims
, and consequently any `rv-only' function that
depends only on these functions will work; e.g. is.constant
, which
depends only on rvnsims
.
The method is.double
is provided for compatibility reasons; this is
needed in a function called by plot.rvsummary
The arithmetic operators and mathematical functions will not work with
rvsummary
objects.
The sims
method returns the quantiles.
Kerman, J. and Gelman, A. (2007). Manipulating and Summarizing Posterior Simulations Using Random Variable Objects. Statistics and Computing 17:3, 235-244.
See also vignette("rv")
.
x <- rvnorm(mean=1:12)
sx <- as.rvsummary(x)
print(sx) # prints the summary of the rvsummary object
#> mean sd 1% 2.5% 25% 50% 75% 97.5% 99% sims
#> [1] 1.019 0.960 -1.208 -0.928 0.373 1.006 1.695 2.839 3.237 2500
#> [2] 2.004 0.985 -0.219 0.061 1.325 1.985 2.665 3.926 4.298 2500
#> [3] 3.010 0.983 0.801 1.138 2.322 2.997 3.685 5.000 5.278 2500
#> [4] 3.990 0.977 1.717 2.086 3.339 3.999 4.637 5.914 6.327 2500
#> [5] 5.017 0.981 2.760 3.093 4.364 5.004 5.682 6.981 7.248 2500
#> [6] 5.989 1.019 3.622 3.999 5.283 5.988 6.685 7.959 8.325 2500
#> [7] 6.999 0.998 4.727 5.036 6.306 7.004 7.693 8.898 9.180 2500
#> [8] 8.022 0.983 5.842 6.168 7.345 8.000 8.690 9.997 10.314 2500
#> [9] 8.994 0.959 6.807 7.153 8.335 8.977 9.612 10.938 11.288 2500
#> [10] 9.998 1.007 7.723 8.063 9.328 9.981 10.649 12.026 12.433 2500
#> [11] 10.991 0.986 8.813 9.090 10.313 11.012 11.653 12.957 13.423 2500
#> [12] 11.973 1.013 9.573 9.929 11.312 11.969 12.656 13.906 14.358 2500
length(sx) # 12
#> [1] 12
dim(sx) # NULL
#> NULL
dim(sx) <- c(3,4) #
dimnames(sx) <- list(1:3, 1:4)
names(sx) <- 1:12 #
print(sx) # prints the names and dimnames as well
#> row col name mean sd 1% 2.5% 25% 50% 75% 97.5%
#> [1,1] 1 1 : 1 1.019 0.960 -1.208 -0.928 0.373 1.006 1.695 2.839
#> [2,1] 2 1 : 2 2.004 0.985 -0.219 0.061 1.325 1.985 2.665 3.926
#> [3,1] 3 1 : 3 3.010 0.983 0.801 1.138 2.322 2.997 3.685 5.000
#> [1,2] 1 2 : 4 3.990 0.977 1.717 2.086 3.339 3.999 4.637 5.914
#> [2,2] 2 2 : 5 5.017 0.981 2.760 3.093 4.364 5.004 5.682 6.981
#> [3,2] 3 2 : 6 5.989 1.019 3.622 3.999 5.283 5.988 6.685 7.959
#> [1,3] 1 3 : 7 6.999 0.998 4.727 5.036 6.306 7.004 7.693 8.898
#> [2,3] 2 3 : 8 8.022 0.983 5.842 6.168 7.345 8.000 8.690 9.997
#> [3,3] 3 3 : 9 8.994 0.959 6.807 7.153 8.335 8.977 9.612 10.938
#> [1,4] 1 4 : 10 9.998 1.007 7.723 8.063 9.328 9.981 10.649 12.026
#> [2,4] 2 4 : 11 10.991 0.986 8.813 9.090 10.313 11.012 11.653 12.957
#> [3,4] 3 4 : 12 11.973 1.013 9.573 9.929 11.312 11.969 12.656 13.906
#> 99% sims
#> [1,1] 3.237 2500
#> [2,1] 4.298 2500
#> [3,1] 5.278 2500
#> [1,2] 6.327 2500
#> [2,2] 7.248 2500
#> [3,2] 8.325 2500
#> [1,3] 9.180 2500
#> [2,3] 10.314 2500
#> [3,3] 11.288 2500
#> [1,4] 12.433 2500
#> [2,4] 13.423 2500
#> [3,4] 14.358 2500