StatXact 9
- Software for small-sample categorical and nonparametric data
New in StatXact 9 | System Requirements | Further Information
"The culmination of ten years of intense research and development,
StatXact 9 is the most complete, up-to-date software package for exact nonparametric
statistical inference on continuous or categorical data in the world. With StatXact
we've covered every mainstream nonparametric procedure, using a full-fledged
GUI implementation under Windows. We have provided at least one solution for
every important nonparametric statistical problem involving either binary, categorical
or continuous data. And, as is our trade-mark, every single one of these solutions
is exact. That is, it remains valid for small, sparse and unbalanced data sets.
All in all you can access over 80 exact procedures through the StatXact pull-down
menus. In addition, we've provided the tools for you to construct your own exact
procedures, or to combine two or more procedures in an optimal fashion through
a suitable choice of scores. When it comes to handling small-sample data, no
other statistical package can even come close to the power and scope of StatXact.
Here are some highlights from the StatXact procedures:"
- Exact tests and p-values for Poisson (person-years) data
- Exact confidence intervals for Poisson (person-years) data
- Shorter exact confidence intervals for binomial data
- Exact confidence intervals for multinomial data
- Conditional (post-hoc) power calculations for stratified linear rank tests
- Exact confidence intervals for ratios and differences of two binomial proportions
- Exact Hodges-Lehmann confidence interval for a shift between two distributions
- Exact goodness-of-fit tests
- Exact Kolmogorov-Smirnov two-sample test
- Exact Friedman and Cochran-Q tests for K related samples
- Exact generalized ANOVA tests
- Exact tests on measures of association
- Exact stratified linear rank tests with stratum-specific scores
- Exact stratified linear rank tests with optimally combined scores
- Exact K-sample tests for censored survival data
(top)
Windows® 7 and Vista® support
Cytel works continuously to give customers a choice of supported operating systems. Cytel's StatXact® and LogXact® are now
both approved for Windows® 7 and Windows® Vista.
R integration
Users can now use R scripts in conjunction with Cytel’s exact statistics software. Computers with R 2.3 or newer may run
existing files or write and run R programs with the R output displayed in StatXact®.
R programs can also run analyses of either StatXact® or LogXact® datasets. All R outputs are displayed in the Cytel Studio
interface (except R plots)
In addition to R, Cytel continues support of its own batch language to for custom analysis and batch calculations
automation.
Methods Tests for two ordered correlated multinomials:
In the Inference for Correlated Categorical Data, Wilcoxon rank sum, Savage scores, Normal scores and Permutation with
general scores tests are added for two ordered multinomials.
In the Power and Sample Size module, exact designs for two independent binomials available:
- Superiority for ratio of proportions (Score test as well as Wald test)
- Noninferiority for ratio of proportions (Score test as well as Wald test)
- Equivalence for ratio of proportions (Score test as well as Wald test)
- The power is computed using exact unconditional type-1 error
Kolmogorov goodness of fit test multinomial distribution:
- Within a 1xc table data, users can now input either expected count or expected probabilities
- Range for the parameter Alpha, the value used in computing the confidence interval, has been changed from [0.001,0.999] to [1.e-6,0.999999]
Only StatXact 9 delivers these capabilities:
- R language support to speed analyses and command of batch jobs
- Stratified Wilcoxon Signed Rank test utilizing case data on paired samples
- Perform test of correlation across single – not bi-variable – correlation
- Compute p-value for the test of trend in multiple outcomes
- Estimate conditional maximum likelihood of trend parameters in correlated data
- Conduct clustered response tests for correlation significance
(top)
System Requirements:
Operating System: Windows 7, Vista, XP
Solaris® (SPARC) 8 and 9 (StatXact® PROCs only)
HP-UX Itanium®-based servers (StatXact® PROCs only)
|
|