Empty Matrix with Dynamic Fertility Model

Hi all,

I am working on a question to replicate the simuelation portion of this code below. I have already generated the Emax matrix.

http://athena.sas.upenn.edu/petra/bergen/sampprog.f90

The code below is what I have. However, I am just getting a matrix full of 0's. Is there something I am doing wrong here?

 simchoice = zeros(1000,20)
%Looping for 1000 women
    for i = 1:1000
         for t = 1:20
            e = normrnd(0 , sqrt(sigma_e_sq),[1000,1]) 
            if t == 1
                N = 0
            else
                N = sum(simchoice(i, 1:(t-1)))
            end
            
            inc = alpha(5) + alpha(6) * t
            simC_0 = inc - (p_n * N)
            simC_1 = inc - (p_n * (N + 1))
            
            simU_0 = simC_0 - (0.5*alpha(1)*simC_0*simC_0) + ((alpha(2) + e)*N) - (alpha(3)*N*N) + (alpha(4)*simC_0* N)
  
        if t < 20
            simV_0 = simU_0 + Emax((N+1), (t+1)) 
        end
 
        simchoice((i+1),t) = 0
        if N <= 7
           simU_1 = simC_1 - (0.5*alpha(1)*simC_1*simC_1) + ((alpha(2) + e)*(N+1)) - (alpha(3)*(N+1)*(N+1)) + (alpha(4)*simC_1* (N+1))
           
           if t < 20
               simV_1 = simU_1 + Emax((N+2), (t+1)) 
           end
           if simV_0 < simV_1
               simchoice((i+1),t) = 1
           end
            
        end
        
        end
        
    end

[Ridwan Alam]

the question is not very clear, sorry. which matrix is empty here?

also, what is simchoice() and what is this statement supposed to do:

simchoice((i+1),t) = 0

[Sarah Schutz]

Hi! I'm sorry. simchoice is a matrix that I wanted to estimate the results of the simulation, but I am always getting a matrix full of zeros. It is supposed to represent the results of a simulation 1000 women about how many children they choose to have

If it helps this is the question I am trying to solve.

http://athena.sas.upenn.edu/petra/bergen/progproblem.pdf

Answer 1

[Ridwan Alam]

Looks like your indexing is off in some places:

simchoice = zeros(1000,20)
%Looping for 1000 women
    for i = 1:1000
         for t = 1:20
            e = normrnd(0 , sqrt(sigma_e_sq),[1000,1]) 
            if t == 1
                N = 0
            else
                N = sum(simchoice(i, 1:(t-1)))
            end
            
            inc = alpha(5) + alpha(6) * t
            simC_0 = inc - (p_n * N)
            simC_1 = inc - (p_n * (N + 1))
            
            simU_0 = simC_0 - (0.5*alpha(1)*simC_0*simC_0) + ((alpha(2) + e)*N) - (alpha(3)*N*N) + (alpha(4)*simC_0* N)
  
        if t < 20
            simV_0 = simU_0 + Emax((N), (t+1)) 
        end
 
        simchoice(i,t) = 0
        if N <= 7
           simU_1 = simC_1 - (0.5*alpha(1)*simC_1*simC_1) + ((alpha(2) + e)*(N+1)) - (alpha(3)*(N+1)*(N+1)) + (alpha(4)*simC_1* (N+1))
           
           if t < 20
               simV_1 = simU_1 + Emax((N+1), (t+1)) 
           end
           if simV_0 < simV_1
               simchoice(i,t) = 1
           end
            
        end
        
        end
        
    end