User Manual for Blossom Statistical Software

ISBN-13: 9781484849200 / Angielski / Miękka / 2013 / 130 str.

Brian S. Cade; Jon D. Richards

Opis

Blossom is an interactive program for making statistical comparisons with distance-function based permutation tests developed by P. W. Mielke, Jr. and colleagues at Colorado State University (Mielke and Berry 2001) and for testing parameters estimated in linear models with permutation procedures developed by B. S. Cade and colleagues at the Fort Collins Science Center, U. S. Geological Survey (known as the Midcontinent Ecological Science Center prior to 2002). This manual is intended to update and replace earlier versions by B. S. Cade and J. D. Richards dated 2000 and 1999 and by W. B. Slauson, B. S. Cade, and J. D. Richards dated 1991 and 1994. We have expanded on material in earlier versions and provide documentation on new routines added since 2000. Routines added since 2000 are: double permutation (HYP/DP) procedures for linear model tests (OLS, LAD regression, and quantile rank score tests) when null models are either implicitly or explicitly constrained through the origin, i.e., no intercept models (Cade 2003, Cade et al. 2005, Cade et al. In press); dropping all but a single zero residual in LAD (and quantile) regression permutation tests of subsets of variables in multiple regression models (Cade 2005, Cade and Richards In press); and computation of all quantile regression estimates (LAD/ QUANT = ALL). In addition, we now offer the option of saving output to a terse formatted file that is useful for summarizing results of multiple simulations (OUTPUT /TERSE or VERBOSE), and the option to store (SAVETEST = file name) the vector of permuted test statistic values from Monte Carlo resampling approximations of probabilities. The computer code has been made more efficient where possible and was compiled with Lahey Fortran 95 to dynamically allocate memory.

Zawartość książki może nie spełniać oczekiwań – reklamacje nie obejmują treści, która mogła nie być redakcyjnie ani merytorycznie opracowana.