autorank vs SAS/STAT
autorank
sherbold.github.ioSAS/STAT
sas.comSide by side
| Figure | autorank | SAS/STAT |
|---|---|---|
| Price from | Free | — |
| Free plan | Yes | — |
| Free trial | — | Yes |
| Open source | — | — |
| Web | — | Yes |
| Windows | — | Yes |
| macOS | — | Yes |
| Linux | Yes | Yes |
| API | Yes | — |
| Self-hosted | Yes | Yes |
| Plans published | 1 | 1 |
| Spec score · Statistical Analysis Software | 8.4 (#23) | 8.9 (#4) |
| Bayesian analysis | Its Bayesian approach uses a Bayesian Signed Rank test and a Region of Practical Equivalence (ROPE). | Yes |
| Deployment | desktop | both |
| Intended users | The project aims to simplify statistical analysis for people who are not statistics experts. | SAS describes SAS/STAT as designed for both specialized and enterprise-wide analytics. |
| Purpose | Autorank is a Python package for comparing and ranking multiple paired populations, such as machine learning algorithms evaluated on multiple data sets. | SAS/STAT provides statistical analysis tools for specialized and enterprise-wide analytics. |
| Regression models | No | Yes |
| Statistical tests | For frequentist analysis, it selects tests based on normality, homogeneity, and the number of populations. | Yes |
| Support | The project invites questions, bug reports, and feature requests through GitHub issues, and welcomes pull requests. | SAS lists SAS/STAT on the SAS Viya platform and SAS/STAT 15.4 among its most recent releases. |
Plans
In their own words
autorank
autorank has no write-up on Specifiction yet; its figures are above.
SAS/STAT
SAS/STAT is statistical analysis software for specialized and enterprise-wide analytics. Its library includes more than 90 procedures for methods such as regression, survey and survival analysis, multiple imputation and Bayesian modeling. Other documented areas include ANOVA, categorical, cluster, multivariate and nonparametric analysis, as well as propensity scores and causal treatment-effect estimation. Methods are available for small data sets, missing values and large data tasks. SAS describes 14 SAS/STAT procedures as multithreaded and makes 12 SAS High-Performance Statistics procedures available for single-machine use. Predictive modeling options include regression, generalized additive, mixed and nonlinear models, and decision trees. Hundreds of statistical graphs and customizable graphics are included. SAS says the software runs on major computing platforms, accesses nearly any data source and integrates into computing environments. A free trial provides access to SAS Viya in a private trial environment. Pricing is available on request. SAS lists SAS/STAT on the SAS Viya platform and SAS/STAT 15.4 among its most recent releases.