@farahamirah Just enter it:

A:= <-delta*D11, 0; 0, D22>;

@mskalsi Here's one that works on both forms

Delta:= e->
     if e::algebraic then diff(e,y,x,x) + diff(e,y,x) + diff(e,y,z,z) + diff(e,x,x) + diff(e,z,x,x)
     else (D[2,1,1] + D[2,1] + D[2,3,3] + D[1,1] + D[3,1,1])(e)
     end if
:

Note that the D form relies on the positions of the function's parameters. It does not care whether you use x, y, or z. On the other hand, diff relies on the names being x, y, z and doesn't care about their order (they don't even have an order).

@mskalsi The D form is only intended to used when f is a function, not an expression.

If you want a differential operator that works on expressions (and won't work on functions), use

Delta:= e-> diff(e,y,x,x) + diff(e,y,x) + diff(e,y,z,z) + diff(e,x,x) + diff(e,z,x,x):

While all these ways of formatting the output are great, I still think that the OP's problem is not using multiplication operators in the input. It'd be great if they'd upload a worksheet instead of posting a picture on tinypic.com.

@jwitte1996 We'll need to see the code of proedure OrderB. But my strong suspicion is that you mistakenly put its output in a print statement.

@Markiyan Hirnyk Thanks, that works for me.

The problem with using Statistics:-MaximumLikelihoodEstimate is that I know of no way within the Statistics package to create a parametrized custom distribution.

Also, what is the variable t?

Edit: Markiyan has cleared up the issue of using MaximumLikelihoodEstimate with parametrized distributions, so once you tell me what t is doing in your formula, we can get to work using it.

@gvelbu12 Yes, that's all obvious. But I don't know what to do to get the bifurcation plot.

@Joe Riel Or better yet

f:= (i,j)-> OrderB(0, i, j+3);

@spalinowy I think that you mean Joe Riel. I didn't write a solution to your problem.

@gvelbu12 So, if I'm interpretting this correctly, you want a plot with tau on the horizontal axis and the accumulation points (i.e., attractors) of the sequence t[k]=x(t[k-1]), t[0]=0 on the vertical axis. Is that right?

The problem is that the range of x(t) is approximately -1..2.6, and Maple can't evaluate the solution further left of -.49883899, probably due to a singularity.

@farahamirah There are only two solutions to D11 = -delta*D11: delta = -1 or D11 = 0. Neither solution leaves both D11 and delta undefined.

@spalinowy You need to use multiplication operators. In 2D input., you can use a space as the operator. 

See ?IntegerRelations,LinearDependency.

@gvelbu12 In that previous Answer of mine, x was simply a variable. In your case x is a function, x(t). The situations don't seem comparable unless you were to fix a specific value of t.