Here's my status: I've been working on fitting routines for ARMA processes to put in src/math/ts/. It turned into more work than the straightforward optimization problem I thought it would be, so I've been working on the details with pen and paper for a few weeks. I should start coding it in a week or so.
-Jason On Sat, Sep 09, 2006 at 01:02:06PM -0700, Ben Pfaff wrote: > I'm quite busy at the moment working on a paper for NSDI 07, > which is going to be the second half of my PhD thesis, so it is > unlikely that I will get much PSPP work done between now and the > Oct. 9 deadline. > > I've accomplished some PSPP work in the past few weeks, but it is > not yet ready for review. I am essentially finished with > revamping the input and output formatting code, as well as > rewriting the documentation. I am now working on adding strict > tests of functionality and comparing PSPP's input and output > against SPSS's. > > I also went off on a significant tangent in which I wrote a bunch > of code for converting among integer and floating-point formats. > This will allow PSPP to read and write little-endian, big-endian, > and VAX-endian integer data, as well as IEEE 754, VAX, and IBM > mainframe hexadecimal floating-point formats, regardless of the > architecture that PSPP actually runs on. I tried to acquire some > system files for testing the VAX and IBM support by posting on > the SPSS newsgroup, but no one replied, so I will probably not > be able to test this support for system files, only for plain > data files. > > -- > Ben Pfaff > email: [EMAIL PROTECTED] > web: http://benpfaff.org > > > _______________________________________________ > pspp-dev mailing list > [email protected] > http://lists.gnu.org/mailman/listinfo/pspp-dev _______________________________________________ pspp-dev mailing list [email protected] http://lists.gnu.org/mailman/listinfo/pspp-dev
