Transformation of coordinate systems

Description

The functions provide mathematical c functions as RMmodels

Usage

RFcalc(model)
R.minus(a, b, factor)
R.plus(a, b, factor)
R.div(a, b, factor)
R.mult(a, b, factor)
R.const(a)
R.c(a, b, c, d, e, f, g, h, i, j, factor)
R.p(proj, new, factor)
R.is(a, is, b)
R.lon()
R.lat()

R.acos(a)
asin(x)
R.asin(a)
atan(x)
R.atan(a)
atan2(y, x)
R.atan2(a, b)
cos(x)
R.cos(a)
sin(x)
R.sin(a)
tan(x)
R.tan(a)
acosh(x)
R.acosh(a)
asinh(x)
R.asinh(a)
atanh(x)
R.atanh(a)
cosh(x)
R.cosh(a)
sinh(x)
R.sinh(a)
tanh(x)
R.tanh(a)
exp(x)
R.exp(a)
log(x)
R.log(a)
expm1(x)
R.expm1(a)
log1p(x)
R.log1p(a)
logb(x)
R.logb(a)
R.exp2(a)
log2(x)
R.log2(a)
R.pow(a, b)
sqrt(x)
R.sqrt(a)
R.hypot(a, b)
R.cbrt(a)
R.ceil(a)
abs(x)
R.fabs(a)
floor(x)
R.floor(a)
R.fmod(a, b)
round(x, ...)
R.round(a)
trunc(x)
R.trunc(a)
R.erf(a)
R.erfc(a)
gamma(x)
R.gamma(a)
lgamma(x)
R.lgamma(a)
R.remainder(a, b)
R.fdim(a, b)
max(...)
R.fmax(a, b)
min(...)
R.fmin(a, b)

Arguments

Details

R.plus

adds two values

R.minus

substracts two values

R.mult

multiplies two values

R.div

devides two values

R.const

defines a constant

R.c

builds a vector

R.is

evaluates equalities and inequalities; note that TRUE is returned if the equality or inequality holds up to a tolerance given by RFoptions()$nugget$tol

R.p

takes a component out of the vector giving the location

R.lon, R.lat

longitudinal and latitudinal coordinate, given in the spherical system, i.e. in radians. (earth system is in degrees).

For the remaining models see the corresponding C functions for their return value. (For any ‘R.model’ type ‘man model’ under Linux.)

Value

Formally, the functions return an object of class RMmodel, except for RFcalc that returns a scalar. Neither vectors nor parentheses are allowed.

Note

The function RFcalc is intended for simple calculations only and it is not excessively tested. Especially, binary operators should be used with caution.

Note that all the functions here are NOT recognized as being positive definite (or negative definite), e.g. cos in R^1:

1. please use the functions given in RMmodels for definite functions (for cos see RMbessel)

2. Using uncapsulated substraction to build up a covariance function is ambiguous, see the example in RMtrend

Author(s)

Martin Schlather, schlather@math.uni-mannheim.de

See Also

RMmodel, RFfctn, RMtrend

Examples

RFoptions(seed=0) ## *ANY* simulation will have the random seed 0; set
##                   RFoptions(seed=NA) to make them all random again

## simple calculation
RFcalc(3 + R.sin(pi/4))

## calculation performed on a field
RFfctn(R.p(1) + R.p(2), 1:3, 1:3) 
RFfctn(10 + R.p(2), 1:3, 1:3) 

## calculate the distances between two vectors
print(RFfctn(R.p(new="iso"), 1:10, 1:10))

## simulation of a non-stationary field where
## anisotropy by a transform the coordinates (x_1^2, x_2^1.5)
x <- seq(0.1, 6, 0.12)
Aniso <- R.c(R.p(1)^2, R.p(2)^1.5)
z <- RFsimulate(RMexp(Aniso=Aniso), x, x)


## calculating norms can be abbreviated:
x <- seq(-5, 5, 5) #0.1)
z2 <- RFsimulate(RMexp() + -40 + exp(0.5 * R.p(new="isotropic")), x, x)
z1 <- RFsimulate(RMexp() + -40 + exp(0.5 * sqrt(R.p(1)^2 + R.p(2)^2)), x, x)
stopifnot(all.equal(z1, z2))
plot(z1)