[R-SIG-Finance] PerformanceAnalytics version 0.9.6 released to CRAN

[Joe W. Byers]

Brian G. Peterson wrote:
> We are pleased to announce the availability on CRAN of 
> PerformanceAnalytics version 0.9.6.
> 
> This is a feature and bugfix release.
> 
> http://cran.r-project.org/src/contrib/Descriptions/PerformanceAnalytics.html
> 
> PerformanceAnalytics is a library of econometric functions for 
> performance and risk analysis. This library aims to aid practitioners 
> and researchers in utilizing the latest research in analysis of 
> non-normal return streams.
> 
> Package: PerformanceAnalytics
> Type: Package
> Title: Econometric tools for performance and risk analysis.
> Version: 0.9.6
> Date: 2007-12-29
> License: GPL
> URL: 
> http://cran.r-project.org/src/contrib/Descriptions/PerformanceAnalytics.html
> URL: http://braverock.com/R/
> 
> New Functions:
>      chart.ECDF
>          Creates an empirical cumulative distribution function (ECDF)
>          overlaid with a cumulative distribution function (CDF)
>          Inspired by:
>          Ruppert, David. 2004.
>          Statistics and Finance, an Introduction.
>          Ch. 2 Fig. 2.5
> 
>      chart.ACF
>      chart.ACFplus
>          Inspired by (and partially ported from) the website:
>          http://www.stat.pitt.edu/stoffer/tsa2/Rcode/acf2.R
>          "here's an R function that will plot the ACF and PACF of a time
>          series at the same time on the SAME SCALE, and it leaves out
>          the zero lag in the ACF [and uses the number of observations
>          as the default]"
>          That description made a lot of sense, so it's implemented here
>          for both the ACF alone and the ACF with the PACF.
> 
>      chart.Regression
>          Uses a scatterplot to display the relationship of returns
>          to a market benchmark.  Fits a linear model and overlays the
>          resulting model.  Also overlays a Loess line for comparison.
> 
>      Return.read
>          Wrapper of 'read.zoo' with some defaults for different
>          date formats and frequencies.
> 
>      Return.Geltner
>          Calculate Geltner liquidity-risk-adjusted return series.
>          David Geltner developed a method to remove estimating/liquidity
>          bias in real estate index returns.  It has since been applied
>          to other return series that show autocorrelation or
>          illiquidity effects. The theory is that by correcting for
>          autocorrelation, you are uncovering a "true" return from series
>          of observed returns that contain illiquidity or manual pricing
>          effects.
> 
>      SmoothingIndex
>          Proposed by Getmansky et al to provide a normalized measure of
>          liquidity risk.  The index will produces a number from zero to
>          one.  A low number indicates low liquidity risk.  A number
>          trending towards one indicates a higher liquidity risk.
> 
>      table.Autocorrelation
>          Produces data table of autocorrelation coefficients rho and
>          corresponding Q(6)-statistic for each column in return series.
> 
>      table.CalendarReturns
>          Returns a table of returns formatted with years in rows, months
>          in columns, and a total column in the last column.
>          For additional columns, annual returns will be appended.
> 
> 
> Significantly Changed Functions:
>      chart.Boxplot
>          Added the ability to more completely control the visual display.
>          Added the ability to render a Tufte-style compact boxplot.
> 
>      chart.Histogram
>          Improved visual display for print-quality graphics
>          Added fits for extra distributions (stable,cauchy,skew-T)
>          Added more control over risk lines
>          Added event lines
> 
>      chart.QQPlot
>          Replaced most internals with port of John Fox's
>          qq.plot from 'car'
>          Now fits arbitrary distributions
>          Allows use of error bands
> 
>      We have made changes throughout the package to allow the
>      risk-free rate to contain a vector of changing rates corresponding
>      with the return series being examined.
> 
>      In addition, we have made more extensive use of the features of the
>      'zoo' package in this release of PerformanceAnalytics, and removed
>      a few external dependencies where those dependencies were minor and
>      easily replicated or ported to this package.  We expect both of
>      these trends to continue in later releases.  Hopefully, we have
>      properly credited the original authors and functions both in our
>      code and in the manual pages.
> 
> Deprecated Functions:
>      rollingCorrelation
>      rollingFunction
>          These functions have been replaced in our code by the use of
>          zoo's 'rollapply' function, and are no longer needed as
>          separate custom functions.
> 
> New Vignettes:
>      We have added as vignettes the presentations we gave on
>      PerformanceAnalytics at the R/RMetrics Conference in Mielesalp
>      in July 2007 and at UseR! 2007 in Ames, Iowa.
> 
> Other:
>      This version of PerformanceAnalytics contains many, many minor
>      improvements and changes.  We added aver 1500 lines of code
>      and comments, and over 1000 lines of documentation.
> 
> We have benefited greatly from feedback and comments from the users of
> PerformanceAnalytics and from R-SIG-Finance.  Please continue to send 
> your questions, comments, and complaints.
> 
> Full details available in the ChangeLog or in the CVS logs in all .R 
> files in the source package.
> 
> Regards,
> 
>      - Brian
> 
Brian,

Great looking package!  I was installing it on my linux server using my 
VNC connection and a terminal window as su root, so I was only using the 
command windows, when I encountered an error with the install.  The 
package requires tclk which will not load unless the display is set 
correctly on the linux terminal shell.  I had to physically log on to 
the linux machine to get it to install.  I am not sure if there is a 
workaround or if you can change the package to only load the tclk 
library as needed, not depend on it.  This is not a major issue for me 
but being able to remotely connect and run a superuser shell is 
convenient.  This will impact anyone that has a cron job that 
automatically updates the R packages on machines.

Happy New Year
Joe