# Release 4.0 of Real Statistics Resource Pack

I am pleased to announce Release 4.0 of the Real Statistics Resource Pack. The new release is now available for free download (Download Resource Pack) for Excel 2007, 2010 and 2013 environments.

The Real Statistics Examples Workbook and Real Statistics Multivariate Examples Workbook have both been updated for capability with the new release and can be downloaded for free.

The Real Statistics website is in the process of being updated to reflect the new features. These changes will be made in the next several days.

The release contains the following new features:

Cluster Analysis

A new Cluster Analysis data analysis tool has been added to the Real Statistics Resource Pack, which implements the K-means++ algorithm. In addition, the following new functions are now available:

CLUST(R1, k, R2): calculates cluster assignment for the data in R1 based on k clusters and the initial cluster assignments in R2. If R2 is omitted then the default initial assignment is used.

ClustAnal(R1, k, iter): runs CLUST(R1, k) iter times and outputs the cluster assignment with the lowest SSE value

CLUSTERS(R1, R2): calculates the cluster assignment for the data in R1 based on the centroids in R2.

CENTROIDS(R1, R2): calculates the centroids for the data in R1 based on the cluster assignment in R2.

SSE_CLUSTER(R1, R2, k) = SSE value for the data in R1 based on k clusters and the cluster assignment in R2

SSE_CENTROID(R1, R2, k) = SSE value for the data in R1 based on the centroids in R2

Multivariate Logistic Regression

A new Multivariate Logistic data analysis tool has been added.

Exponential Regression

A new Exponential Regression data analysis tool has been added which implements the non-linear algorithms for performing exponential regression.

Reorganization of the data analysis tools

Access to the (Multiple) Linear regression, (Binary) Logistic Regression, Exponential Regression and Multinomial Logistic Regression tools is now available via a Regression dialog box. Additional regression algorithms will be added in the future.

Access to Hotelling’s T-square test, MANOVA, Factor Analysis and Cluster Analysis is now available via a Multivariate Analyses dialog box. Additional multivariate capabilities will be added in the future.

The Mann-Whitney, Wilcoxon Signed-Ranks and Kruskal-Wallis data analyses can now be accessed from the Non-parametric Tests (previously called Additional Non-parametric Tests).

Iterative proportional fitting procedures (IPFP)

The following array functions have been added:

IPFP2(R1): outputs the results of the IPFP algorithm for two-way contingency tables

IPFP3(R1, R2): outputs the results of the IPFP algorithm for three-way contingency tables

Minor changes and bug fixes

• The default Cutoff value for Item Discrimination option of the Reliability Procedures data analysis tool has been changed to .27.
• The non-central chi-square functions NCHISQ_DIST, NCHISQ_INV and NCHISQ_NCP now accept non-integer values for the degrees of freedom.
• An error in the TiesCorrection function for paired Wilcoxon Signed-Ranks test when there is missing data has now been corrected.
• Bugs in the ExpCoeffSRANK_EXACT and SRANKPair_EXACT functions have now been fixed.

I had hope to definitively fix the problem with the Add-Ins ribbon disappearing from time to time in Excel 2013. This ribbon provides an alternative way to access the Real Statistics data analysis tools for those who prefer not to use the keyboard approach Ctrl-m.

For those of you who are experiencing this problem, you can use the following workaround

(1) Press Alt-TI. On the dialog box that appears unclick the Analysis ToolPak and ToolPak – VBA add-ins and click on the OK button.
(2) Close Excel and reopen Excel
(3) Press Alt-TI. On the dialog box that appears click the Analysis ToolPak and ToolPak – VBA add-ins and click on the OK button.

This entry was posted in Announcement, New Release. Bookmark the permalink.

### One Response to Release 4.0 of Real Statistics Resource Pack

1. ndm says:

Release 4.0 is *superlative*. Thank you, Zaiontz!!!!