| |
|
PROGRAM OPTIMIZEFinal MattersContact me for three test datasets. One is an arbitrary set of sequences of varying lengths, with a lookup table of interelement distances. This is OPTEST.SEQ and OPTEST.IED. RANSEQ1 is a bigger random sequence dataset. The two datasets are continuous variable datasets, both from data sent to me by David Weakliem. XXXXWAGE.DAT is data on wage changes in 21 industries over a 40 year period. It has one variable (wage change), which has been differenced to remove trends. XXXXW12 has four variables, all of them standardized over the grand means for all cases and years (and therefore retaining trends). There are 21 sequences and 36 years of data. The variables are corporate profits before tax, corporate capital construction allowance, # of FTE employees, and average hourly wages of production workers. Do NOT attempt to use these data for any purpose other than familiarization with this program: first because they belong not to me but to Professor Weakliem and second because I did the preparation work on them four years ago and my notes are a little unclear about what exactly I did. (I could be wrong, for example, about the identity of one of the variables.) OPTIMIZE was compiled by Borland C++ compilers, although at present the program does not use any C++ attributes, that is, it isn't yet object-oriented. The *.BGI and *.CHR files in the OPTDOS.ZIP are necessary for the graphics to work correctly under MS-DOS. So, to install, create a subdirectory and copy the files into it. Then do your analysis in this directory or, if you work on virtual disks all the time like me, make sure that this directory lies on your specified path if you don't copy everything to wherever you are analyzing. The code itself is found at http://www.src.uchicago.edu/users/abbot/om.html. It also contains a read.me file. Please notify me about modifications you have tried on your own. The program was written by Grant Prellwitz of Prellwitz Computing Services on a draft of programs originally written by me. Future modules will include a subsequence algorithm, for finding multiple subsequence alignments. Please send all comments, reports of bugs, suggestions of modifications, etc. to Andrew AbbottDepartment of Sociology University of Chicago 1126 E 59th St. Chicago IL 60637 a-abbott@uchicago.edu Please note that, while this program is freeware, it is still copyrighted. Thanks. OPTIMIZE © 1997 Prellwitz Computing Services. All rights reserved. |
Limits
Back to Top
|
 |
|||
|
Contact Information: |
|||