######################################################################
#  On arithmetics operating on distributions in package "distr"       
######################################################################


Attention:

Arithmetics on distribution objects are understood as operations on corresponding
random variables (r.v.'s) and _not_ on distribution functions or densities; 
e.g.

    sin( Norm() + 3 * Norm() ) + 2
    
returns a distribution object representing the distribution of the r.v. 

     sin(X+3*Y)+2
     
where X and Y are r.v.s i.i.d. N(0,1).


Binary operators like "+", "-" would loose their elegant calling
 e1 + e2 if they had to be called with an extra argument controlling
their accuracy. Therefore, this accuracy is controlled by global options.
These options are inspected and set by distroptions(), getdistrOption(),
see ?distroptions

The operands of these binary operators are assumed stochastically independent.
So for a distribution object X, the expressions X+X and 2*X are _not_
equivalent. 
More precisely, whenever in an expression, the same symbol for an object 
occurs more than once, for every instance a new independent distribution
is generated.  
If you want to realize the distribution of the random variable
X+X where (twice identically the same random variable) therefore always use
2*X.

At several instances (in particular for functions from group Math like 
sin(), exp()) new distributions are generated by means of RtoDPQ, RtoDPQ.d,
i.e. after simulating a large number of random variables, the slots
d, p, q are filled with stochastic estimates. So don't be surprised if 
they will change from call to call.
