To calculate portfolio returns
3 visualizaciones (últimos 30 días)
Mostrar comentarios más antiguos
Qian cao
el 16 de Nov. de 2015
Respondida: Guillaume
el 16 de Nov. de 2015
Hi, I have a question regarding the matrix multiplication with NaN cells in the matrix.
The two matrix are
W1= [0.1 0 0.9; 0 1/3 2/3; 1/3 1/3 1/3;0.2 0.4 0.4];
r=[0.2 0.1 NaN; -1 0 NaN; NaN 0.2 0.2; NaN 0.1 0];
The rows represent the time series. What I need is to get the vector R(t) which is the multiplication of W1(t-1)*r(t). That is R(2)=W1(1)*r(2)=[0.1 0 0.9]*[-1 0 NaN]=-0.1, R(3)=W1(2)*r(3)=[0 1/3 2/3]*[NaN 0.2 0.2]=0.2 and vice verse .I did it in the following way but the results do not show up as expected....Note: treat the NaN values as zeros:
for t=1:size(W1,1)
R(t,1)=W1(t-1,:)*returnm(t,:)';
end
Thank you very much for your help. I am waiting for that.
0 comentarios
Respuesta aceptada
Guillaume
el 16 de Nov. de 2015
If you want to treat the NaNs as zero, make them zero:
r(isnan(r)) = 0;
Your calculation is then simply
R = [NaN; sum(W1(1:end-1, :) .* r(2:end, :), 2)] %no loop needed
I've put NaN for R(1) since you haven't specified in your description what it should be. Personally, I would simply change the offsets of R (i.e. have R(1) = sum(W1(1) .* r(2)))
0 comentarios
Más respuestas (0)
Ver también
Categorías
Más información sobre Portfolio Optimization and Asset Allocation en Help Center y File Exchange.
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!