VVm {UCS} | R Documentation |
Computes the variances of the frequency spectrum and conditional parameter distribution of a LNRE model (Baayen, 2001) at sample size N.
VVm(model, m, N, rho=1, relative=FALSE, lower=TRUE)
model |
an object of class "zm" or "fzm" ,
representing a Zipf-Mandelbrot (ZM) or finite Zipf-Mandelbrot (fZM)
LNRE model |
m |
a vector of positive integers, representing frequency ranks |
N |
a vector of positive integers, representing sample sizes;
either m or N should be a single number |
rho |
a vector of numbers in the range [0,1]. If
length(rho) > 1 , both m and N should be single
numbers. See below for details. |
relative |
if TRUE , computes variances for the relative
conditional parameter distribution (see below for details). May
only be used when rho is specified. |
lower |
if rho is specified, controls whether variances
are computed for the lower or for the upper conditional parameter
distribution |
The variance V[V_m(N)] is computed according to Baayen (2001, 120f).
When rho
is specified, the variances of the conditional
parameter distribution V[V_{m,ρ}(N)] or the corresponding
proportions V[R_{m,ρ}(N)] are returned, depending on the
value of relative
. With lower=FALSE
, computes variances
for the upper conditional parameter distribution
V[V_{m,>ρ}(N)] or proportion V[R_{m,>ρ}(N)].
See Evert (2004, Ch. 4) for details.
The EVm
help page provides more information about
V_m(N), V_{m,ρ}(N), R_{m,ρ}(N),
V_{m,>ρ}(N) and R_{m,>ρ}(N).
Note that this function does not compute variances for the relative frequency spectrum (V[V_m(N)/V(N)]) or the ratio between consecutive spectrum elements (V[V_{m+1}(N)/V_m(N)]).
a numeric vector of appropriate length (determined either by m
,
N
, or rho
)
Baayen, R. Harald (2001). Word Frequency Distributions. Kluwer, Dordrecht.
Evert, Stefan (2004). The Statistics of Word Cooccurrences: Word Pairs and Collocations. PhD Thesis, IMS, University of Stuttgart.