3153 Reputation

18 Badges

14 years, 200 days

MaplePrimes Activity

These are questions asked by MrMarc

One thing I think is a bit strange is the notion that the loop counter
is assigned a value after the loop is done. Why is that ?

for example:

for i to 10 do
end do;

The above code will simply print 1..10 but then if you simply run:


then 11 will be printed out. How often do you actually need i to be assigned
a value after the loop is done?

I think it would have been much better if i is unassigned when the loop is done

I want to show algebraically that the expected correlation is
decreasing in the tails of the distribution ie that the correlation coefficient
breaksdown when we have large moves. How can this be done?

I have been watching a demonstration on youtube about LP and Interior Point Methods:

He is solving a LP by using a barrier function and has a very nice simulation
in the end where the solution converges to he optimal point and mu goes to zero.
I tried to replicate it in Maple but I am strugeling.
I have attached an worksheet what I have done so far:

what I dont understand is the purpose of the second order cone constraint.

The definition is: The first element must be at least as large as the Euclidean
norm of the remaining elements. The euclidean norm for R^3 is the distance from
origo to the point [x,y,z] due to pythagores theorm ie A^2+B^2=C^2 where
C is the distance from origio to the point[x,y,z]

Again for R^3 then we have [x,y,z] then x > the euclidean norm for (y,z) which
means that x> sqrt(y^2+z^2...

If I use a binary {0,1} linear program the cardinal constraint ie control the
number of 1's (long positions in portfolio) works beautiful. See attached worksheet.

The problem starts when I convert the problem into a constrained integer {-1,0,1}
linear problem where -1= short position, 0=no position and 1=long position.

In the first example when we add the constraint  add(w[i], i = 1 .. NC) = 4

First 8 9 10 11 12 13 14 Last Page 10 of 35