Publications by DomPazz
Learning R: Project 1, Part 2
So it’s been a week since I started down this path. I worked most of this out over last weekend, went to a conference, had hectic week at work, and then realized I lost my work. Gah.I’ll be posting my general thoughts on R later. Mostly it seems to be a neat language. Lots of ways to do things. The ability to create output...
4894 sym 8 img 2 tbl
First thoughts on R
Having worked just a little with R, I have some first impressions to share. I’ll give you some links to resources I found helpful with writing the previous project. First, the documentation is not very good. I struggled on previous attempts to figure things out. I still find it crap shoot when I Google, looking for an answer. ...
3492 sym 6 img
Matrix Performance in R
I’ve been working on an example of the new Graph Template Language from SAS. As I don’t have direct access to SAS 9.2, I’ve been developing via email with a friend that does.In the meantime, I thought I would start to investigate some of the performance properties of R. I work in the financial risk industry and I often find ...
3804 sym
Matrix Package Doodling
Trying not to fall into Thanksgiving Day, football, coma. So I started looking at the Matrix package.Started out by changing my code from before to create a matrix using the Matrix() function from the Matrix package.n = 4000c = Matrix(.9,n,n)for(i in 1:n){ c[i,i] = 1;}That took… a LONG time. What the hell? Reading the intro...
3499 sym 2 img
Dealing with Non-Positive Definite Matrices in R
Last time we looked at the Matrix package and dug a little into the chol(), Cholesky Decomposition, function. I noted that often in finance we do not have a positive definite (PD) matrix. The chol() function in both the Base and Matrix package requires a PD matrix. In this post, we will look at why we don’t always have a PD mat...
6276 sym
Non-PD Matrices in R, Cont.
Let me preface this post by saying I am getting frustrated with R. The syntax is not intuitive and the performance for matrix operations is slow. Using Octave, a free Matlab clone, I can get over 6 Gflops on things that R is doing at less than 2. After this post, I will focus on the statistical functions of R and will do any matrix opera...
6216 sym 17 tbl
I may have been hasty…
I think one of the real reasons that I haven’t liked R is the default interface blows (sucks, whatever). I just discovered the Eclipse plugin StatET. This things rules. Contextual help, completion, object browser, data browser, etc. There is even a debugger that I haven’t tried yet. Thank you WalWare!Still not going to u...
850 sym
Stat-ET
Working on a new post and I’m starting to enjoy Stat-ET. Here’s a screen shot (click to view full size)Notice the object browser. Want to know what functions a package has? Check there. Want to see what is in that data frame, select it in the browser and hit F3. Graphical output is on the right, console on the bottom. Wa...
930 sym 2 img
Hello Mr. Bernake…
I was playing around tonight and came across something that looked odd. Using the importSeries() created before, I grabbed dividend adjusted returns for SHY, IEF, and TLT (iShares Short, Medium, and Long Maturity Treasury ETFs respectively). I was looking at efficient frontiers (more on that later) and saw that the IEF was very do...
1518 sym 1 tbl
Portfolio Optimization in R, Part 1
I briefly mentioned in my last post; that I was fooling around with portfolio optimization in R. This post will the first in a series on the topic of portfolio optimization. Please note, nothing I am about to say should be taken as advice for investing. These results are based on prior observed returns and the future rarely mimics the past....
7456 sym 2 img 1 tbl