Developer: Cytel Inc.
Latest Release: StatXact 10, LogXact 10
Operating System: Windows
Cytel offers a combination of statistical analysis tools (including Egret, StatXact, LogXact and ToxTools) and clinical trial design, similution and implementation software (including SiZ).
Cytel develops award-winning statistical data analysis packages, and trusted solutions for clinical trial planning and operations support.
Rigorously tested and continuously validated in practice, Cytel software is used worldwide in over one thousand pharmaceutical, biotech and medical device companies. It is also popular at leading academic and independent medical research institutes.
Health authorities in North America, Europe and Japan regularly refer to our software products.
StatXact 10: 150 Tests and Procedures for Exact Inference
Exact p-values are often considered the gold-standard for measuring the strength of a statistical finding. However, computing p-values while evaluating large or unbalanced data sets, can raise an assortment of challenges for statisticians.
StatXact offers an expansive toolkit of non-parametric exact and Monte Carlo methods for accurate hypothesis testing. Its range of application extends across the natural and social sciences. StatXact has been instrumental in decisive findings across in all conceivable fields of study, including: economics, population, public health, the life sciences, clinical analysis, natural resources, law, governmental policies and defense.
The Advantages of StatXact
StatXact offers more tests and procedures for exact inference and power analysis than any other software package on the market. Fast and accurate inferences are compiled on an easy-to-use interface, drawing on complex algorithms created by Cytel’s team of statistical experts.
Our experts are constantly developing new types of exact tests, broadening the selection of calculations available to our users. Many of these methods are only available to StatXact users, and have received acclaim for their performance from both industry and academia.
LogXact 10: Exact Inference for Logistic Regression
The complexity of conducting regression analysis over multiple covariates is well-documented. The challenge only intensifies when coupled with small sample sizes or missing data sets. LogXact aims to provide simple and accurate solutions for such difficulties.
LogXact can handle many varieties of response data including continuous and binary, polytonomous, count, and missing data. Users of the software can be confident of in their results derived from LogXact’s advanced regression techniques.
The Advantages of LogXact 10
LogXact 10 offers pioneering methods in exact inference and regression modeling to provide rapid and accurate analysis. Powerful algorithms are built to analyze stratified and unstratified data sets of all sizes. Cytel-developed algorithms are carefully tested, highly validated in practice, and compliant the guidance found in the FDA’s CFR Part 11.
LogXact continues to set the standard in all facets of regression modeling using exact inference. It is the first to offer the widely acclaimed PMLE procedure that minimizes separation bias while exhibiting significantly lower error rates compared to typical maximum likelihood estimators.
The Toolkit: LogXact performs regression analysis for continuous, binary, polytonomous and count data. It can also apply advanced regression techniques to data sets with missing values.
Missing Covariates: Only in LogXact can users accurately fit general linear models in cases of missing categorical covariates (models include Logit, Probit, CLoglog, Poisson and Normal.)
Large Data Sets: LogXact provides options to handle large data sets using exact methods, Monte Carlo sampling and Markhov Chain Monte Carlo sampling. A useful ‘Exploration Mode’ allows users to specify parameters to build networks that satisfy analysis needs.
New Developments: LogXact has recently added several features to its world-class toolkit including Firth’s PMLE procedure; the calculation of mid-p corrected confidence intervals for a variety of models; best subset selection in binary logistic regression; the force inclusion of variable to the best subsets; and profile likelihood confidence intervals for parameters of binary logistic regression.