hold on is plotting strange lines to 0 when used for this code. Any ideas why?

gamma=1; alpha=0; beta=0; N=100; x0=50;

xn=zeros(1,N); xn(1)=x0;

realisations=1;

for i=1:realisations

      hold on

      for j=1:N-1

          pn = (((gamma*xn(j)*(1-alpha))+(N-xn(j))*beta)/(gamma*xn(j)*(1-alpha)+...
              (N-xn(j))*beta+(N-xn(j))*(1-beta)+gamma*xn(j)*alpha))*(xn(j)<=N);

          xn(j+1)=binornd(N,pn);

          plot(xn)

          axis([-10 N+10 -10 N+10])
      end
  end

Which produces a plot like this:

I was hoping someone could help as to why it plots these lines to 0.

Thanks

[James McCusker]

Sorry I should have changed the value of realisations in the code provided to something greater than 1 - I intended to plot multiple realisations of xn in the same plot.

Answer 1

[Walter Roberson]

gamma=1; alpha=0; beta=0; N=100; x0=50;

xn=zeros(1,N); xn(1)=x0;

realisations=1;

for i=1:realisations

        hold on

        for j=1:N-1

            pn = (((gamma*xn(j)*(1-alpha))+(N-xn(j))*beta)/(gamma*xn(j)*(1-alpha)+...
                (N-xn(j))*beta+(N-xn(j))*(1-beta)+gamma*xn(j)*alpha))*(xn(j)<=N);

            xn(j+1)=binornd(N,pn);

        end
    end

    plot(xn)

    axis([-10 N+10 -10 N+10])

[James McCusker]

Sorry I forgot to mention that I need multiple realisations on the same plot hence using hold on and plot inside the for loop.

[Walter Roberson]

gamma=1; alpha=0; beta=0; N=100; x0=50;

xn=zeros(1,N); xn(1)=x0;

realisations=1;

for i=1:realisations

          hold on

          for j=1:N-1

              pn = (((gamma*xn(j)*(1-alpha))+(N-xn(j))*beta)/(gamma*xn(j)*(1-alpha)+...
                  (N-xn(j))*beta+(N-xn(j))*(1-beta)+gamma*xn(j)*alpha))*(xn(j)<=N);

              xn(j+1)=binornd(N,pn);

          end

         plot(xn)
      end

      hold off
      axis([-10 N+10 -10 N+10])

[James McCusker]

Thank you so much! Do you know why that solved the problem? Just curious

[Walter Roberson]

You were redrawing at each step with the complete xn, including all of the 0's you had initialized with. After j = 1 only xn(1:2) had non-zero values so it had to drawn a return to 0 to plot all of xn after the second value. After j = 2, only xn(1:3) had non-zero values so it had to draw a return to 0 after the third value to plot all xn... and so on.