1
$\begingroup$

I am looking to run the optimization problem. I already finish my optimization code as below.

expr = {x*h + 2*g*y, x + y*g/2 > B, 0 < x < 10, 0 < y < 10};
tab = Table[{B, x, h, g} /. #2 & @@ NMaximize[expr, {x, y,}], {B, 1, 
    3, 0.1}];
TableForm[tab, TableHeadings -> {None,    {"B", "x", "h", "g", "optimal point"}}]

However, the values of h and g should come from exponential distributed random number generator.

In matlab, I can do:

 h_array=exprnd(1,100);  %generate 100 exponential distributed random number
 g_array=exprnd(1,100);  %generate 100 exponential distributed random number
 for i=1:100
        h=h_array(i);
        g=g_array(i); 
        i++;
        <then run the similiar optimization code  in matlab as code given above>
 end;

How do I reproduce the same thing with my optimization code given above in Mathematica?

$\endgroup$
1
  • $\begingroup$ Look up RandomVariate[] and ExponentialDistribution[]. $\endgroup$
    – J. M.'s torpor
    Aug 5 '15 at 18:24
2
$\begingroup$

The random functionality within Mathematica is all of a pattern:

RandomFunction[range, outputStructure]

where range depends on what RandomFunction you are using, e.g. for RandomInteger and RandomReal it is {min, max} and they both default to {0,1} if no min/max are supplied. The outputStructure tells the RandomFunction how many random numbers you want and in what layout. For example, to generate 100 pairs of integers between -1 and 1, I would use

RandomInteger[{-1, 1}, {100, 2}]

For your question, you are looking for RandomVariate with a "range" of ExponentialDistribution. Conceivably you could then do this

g = RandomVariate[ExponentialDistribution[1], 100];
h = RandomVariate[ExponentialDistribution[1], 100];

But, I would probably use it like this

{g, h} = RandomVariate[ExponentialDistribution[1], {2, 100}];
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.