New features in XLSTAT 2015

XLSTAT 2015 adds a number of new features:

  • XLSTAT-Pro :
    • A major breakthrough in statistical computing speed has begun, with the first function being GPU-enabled. Running Monte Carlo simulations for nonparametric tests can now be done on NVIDIA graphic cards, faster than ever. From now on, XLSTAT features will be gradually GPU-upgraded. XLSTAT version 2015.5 is compatible with the new Excel 2016 (Windows only, Mac to be announced later).
    • Added the K-Nearest Neighbors (KNN) classification algorithm
    • Cochran Q non-parametric test now includes exact and Monte Carlo approximation options to compute p-values
    • ANOVA - the Benjamini-Hochberg False Discovery Rate penalization method is now available among the multiple pairwise comparisons options.
    • Several exciting options have been added to the TURF feature, including outputs summarizing contributions to the REACH.
    • A brand new feature allows you to compare an observed variance to a theoretical one.
    • Scatter plots: confidence ellipses can now be drawn based on Chi-square or Fisher statistics.
    • K-Nearest Neighbors: it is now possible to mix both qualitative and quantitative predictors for a more accurate supervised machine learning.
    • Quantile regression: confidence intervals can now be added to parameter estimations.
    • Correspondence analysis: the Hellinger distance can now be used instead of the Chi-square distance.
    • Mixed models: computational speed has been significantly improved.
    • K Nearest Neighbors: enjoy broader machine learning possibilities by studying quantitative dependent variables.
    • ANOVA: it is now possible to automatically select the same set of factors for multiple comparisons if several dependent variables are selected in multi-factor models.
    • Ternary diagrams: a new option allows to update the diagram automatically when you change the input values.
    • RV coefficient: an easy way to evaluate the similarity between two data matrices. 
    • Analyzing data-Correspondence analysis: draw confidence circles around row and column categories to identify those that contribute to the hypothesis of independence.
    • Modeling data-ANOVA: the Games-Howell procedure will let you run multiple pairwise comparisons without having to worry about unequal variances.
    • Naive Bayes now enriches your supervised machine learning algorithms collection. Naive Bayes assumes independence among predictors. It is fast and known to be very accurate in many situations.
    • Our PLS Path Modeling tool now includes an interface specially adapted to Marketing and Satisfaction analysis. Among the new available options, you will be able to evaluate the impact of modifying brand image, quality or value on customer satisfaction. PLS-Path Modeling is part of the XLSTAT-Marketing and XLSTAT-Psy solutions.
  • XLSTAT-MX:
    • The CATA feature has been considerably improved. It now includes a test of independence associated to the correspondence analysis and an option to filter out assessors or attributes.
    • The CATA feature now includes the Hellinger distance option to limit the influence of under-represented attributes in the correspondence analysis. XLSTAT is one of the only statistical software to introduce this option.
    • The CATA feature now displays mean drops charts that allow an easier identification of attributes to withdraw or to include in your products, as well as those that are nice to have.
    • The CATA feature now includes a plot summarizing comparisons of elicitation rates for each product and the ideal product.
  • XLSTAT-3DPlot :
    • You may now intuitively filter out data on the plots to enjoy deeper and more precise data mining processes.
    • Enjoy perspective in your data with a renewed, fresh and modern interface.
  • XLSTAT-OMICs:
    • Comparing heatmaps of different datasets is now possible thanks to the user-defined min and max color calibration option.
  • XLSTAT-Time:
    • you may now run Unit Root/stationarity tests on heteroscedastic data, thanks to the Phillip-Perron test option.
    • A new feature allows to produce elegant time series charts in a few clicks and manipulate them in real time.
    • Time series transformation: decompose your time series into three parts: trend, seasonal and random.
    • The new cointegration of time series feature will help you test the similarity of trends on the long run according to Johansen’s approach.
  • XLSTAT-Life
    • Deming regression: the computation of confidence intervals now match the EP09-A3 CLSI standard.
  • XLSTAT-Conjoint
    • MaxDiff designs may now be generated in a new format (combinations in columns and respondents in rows).
    • MaxDiff analysis can now be done using the very common multinomial logit model.

The changes from version 2013 to 2014, and those added in 2014, can be found here.

About KCS

Kovach Computing Services (KCS) was founded in 1993 by Dr. Warren Kovach. The company specializes in the development and marketing of inexpensive and easy-to-use statistical software for scientists, as well as in data analysis consulting.

Mailing list Join our mailing list

Home | Order | MVSP | Oriana | XLStat
QDA Miner | Accent Composer | Stats Books
Stats Links | Anglesey

 

Like us on Facebook Facebook

Get in Touch

  • Email:
    sales@kovcomp.com
  • Address:
    85 Nant y Felin
    Pentraeth, Isle of Anglesey
    LL75 8UY
    United Kingdom
  • Phone:
    (UK): 01248-450414
    (Intl.): +44-1248-450414