Storing results from a for-loop in a vector of zeros

1 view (last 30 days)
I have the code Backtesting_10day_II (here attached).
From line 23-26 I created 4 vectors (of zeros) to pre-allocate the results from the for-loop, each of these results is the value of a portfolio calculated every 10 days (V_min_var, V_eff_var, V_min_ES, V_eff_ES).
Now in the for loop: i goes from 1001 until 3721, there are 273 iterations (n= 3740 and i=1001:10:n-19), so thinking logically there should be only 274 results per portfolio (274 values per portfolio), which means that each pre-allocating vector should be of size 274x1.
After I run the program I checked the pre-allocating vectors and I found that they were size 3731x1, the majority of the numbers were zeros (zeros from row 1 to 1000,then actual values at rows 1001,1011,1021,1031,etc until 3731, but everything in between is zero). I don't understand this because I predefined these vectors to be of size 274x1.
Why do I have so many zeros in between my results? and how can I fix this? I only need the results not the zeros in between.
Thanks!

Answers (1)

Iain
Iain on 20 Aug 2014
Indexing in matlab is done from a base of one.
So,
A(3479) = 1;
will change the 3479th element of A to 1.
You need to change how you are indexing your variables with the iteration number, rather than the loop variable. - How you manage that is up to you.
  10 Comments
Iain
Iain on 21 Aug 2014
You are confusing "value" for "index".
You must always address vectors with an index. An index can be related to a value by a simple algorithm (eg (i-991)/10 ), or by finding a value's position in a list (find(x == 45,1))
civs
civs on 21 Aug 2014
Ok, but how would it work for my problem here...? find(x==1001:10:3731)
I need a concrete answer. It doesn't need to be a sophisticated solution, just a simple solution will do.

Sign in to comment.

Categories

Find more on Portfolio Optimization and Asset Allocation in Help Center and File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!