Analysis of Financial Time Series
Dear Brian:
What do you think of "http://fints.r-forge.r-project.org" and the
attached "FinTS Project Work Plan.xls"?
I think I should next do the following:
1. Fix the obvious display problems with this web site.
2. What do you think about posting questions to R Wiki, then
asking package maintainers (and later R-SIG Finance) to help resolve
questions? That should stimulate a discussion whose resolution should
be reasonably well documented and accessible to others. I plan to try
that with an ARCH question from Tsay's chapter 3.
3. Upload a new version of FinTS to CRAN.
4. When that's available, issue a progress report to
R-SIG-Finance, inviting responses via R Wiki.
Best Wishes,
Spencer Graves
Brian G. Peterson wrote:
Spencer Graves wrote:
I'm currently developing an R companion to Ruey Tsay (2005) Analysis of Financial Time Series, 2nd ed. (Wiley), and I'd be pleased to have help.
Spencer, I am very pleased to hear that you are making progress in this, as it is important work that many students, professionals, and researchers will benefit from. I'm wondering if you've made a table of the examples or chapters and the specific topics or techniques employed in each. I think that there are many people on this list who have expertise in specific techniques, as you suggest below. I think that categorizing the examples by goal or technique could help enlist aid from this group, as individuals could choose a small number of examples that use a technique that they are are either already expert in in R, or a technique which they wish to learn in greater detail. I'm traveling right now, and my copy of Tsay is at home, but I would be happy to help construct such a cross-reference table if it doesn't already exist. Regards, - Brian
A preliminary version "FinTS 0.1-17" is now available on CRAN.
A slightly newer version can be installed from R-Forge via
'install.packages("FinTS",repos="http://r-forge.r-project.org")'.
The source code is available via "svn checkout
svn://svn.r-forge.r-project.org/svnroot/fints".
The current versions contain all the data sets used in the text
as documented R objects plus script files to generate nearly all the
analyses in chapters 1 and 2. It is therefore a great help, I
believe, for anyone reading this book.
I could use volunteers to help me complete the package. So
far, I've found R functions to reproduce nearly all the examples,
figures and tables in the book. However, I don't use them routinely,
and it is taking me considerable time to find what is available, to
decide which of the available functions seem most appropriate for
each application, and figure out how to use it so I get results
reasonably close to those in the book. For example, chapter 3
discusses "Conditional Heteroscedastic Models", including ARCH,
GARCH, EGARCH, CHARMA, random coefficient autoregressive models, and
stochastic volatility models. Other chapters discuss nonlinear time
series, high frequency and continuous time models, extreme values,
multivariate time series, Kalman filtering, and MCMC. I can follow
the math, but I have not used many of these models myself.
If you are interested in helping with this project, please let
me know.
-------------- next part -------------- A non-text attachment was scrubbed... Name: FinTS Project Work Plan .xls Type: application/vnd.ms-excel Size: 13824 bytes Desc: not available Url : https://stat.ethz.ch/pipermail/r-sig-finance/attachments/20071221/6bcc2702/attachment.xls