Re: Best Way to Import Multple Return Series?
- From: "Markus" <no@xxxxxxxxx>
- Date: Tue, 13 Sep 2005 09:56:01 +0300
"Sean B" <shady757@xxxxxxxxxxx> wrote in message
news:ef13d15.-1@xxxxxxxxxxxxxxxxxxx
> Hi, this is pretty basic, but I am a brand new Matlab user. Say I
> have a portfolio of 12 different assets, with 10 years of monthly
> retrun data that I want to import from Excel... I am going to be
> calcing things like avg, stdev, covar and looking at effecient
> frontier (portopt) etc...
>
> What is the best way to do this? I would think with the import
> wizzard, but I am having problems. Should I import them in as a
> single 12x120 matrix called returns? If so what about my date
> column? and my column headings? Those are text?? Do I need the
> date column?
>
> Does the Excel file need to have special formatting? CSV etc...?
>
> Should I be importing them with a new variable for each asset?
> asset1 asset2 etc...? If i do that how do i then go about creating
> the covariance matrix and running the optimization on all these
> seperate matrices?
>
> Should I be importing them all as seperate variables then
> concatenating them into a big 12x120 matrix for doing things like the
> portfolio optimization? again what about the date column in this
> case? leave it out?
>
> Thaks for any advice anyone has....
My advice would be to go with your instincts. As long as you
have your data safe in Excel workbook you can start exploring
Matlab. It's not a bad idea to keep your data in separate vectors
first. Although in time you will learn to process it in one big matrix.
We all do.
Questions like these are the most difficult to answer. If you have some
specific problems I'm sure we can be more of a help.
.
- References:
- Best Way to Import Multple Return Series?
- From: Sean B
- Best Way to Import Multple Return Series?
- Prev by Date: Re: Help with some optimizations please
- Next by Date: Re: problem in importing xls file
- Previous by thread: Best Way to Import Multple Return Series?
- Next by thread: Help with some optimizations please
- Index(es):
Relevant Pages
|