I am pleased to announce Release 7.9 of the Real Statistics Resource Pack. The new release is now available for free download at Download Resource Pack for Excel 2010, 2013, 2016, 2019 and 365 in both Windows and Mac environments.
The examples workbooks will be updated shortly for compatibility with the new release. Over the course of the next few days, the website will also be updated for compatibility with the new release.
If you are getting value from the Real Statistics website or software, I would appreciate your donation to help offset the costs of the website by going to Please Donate.
The following is an overview of the new features in Release 7.9:
Descriptive Statistics for Frequency Tables
The following worksheet functions have been added where R1 and R2 define a frequency table.
MEDIAN_FREQ(R1, R2) = median of the data in the frequency table
MAD_FREQ(R1, R2) = median absolute deviation (MAD) of the data in the frequency table
SMALL_FREQ(R1, R2, k) = the kth smallest element of the data in the frequency table
PERCENTILE_FREQ(R1, R2, p, b) = the pth percentile of the data in the frequency table
QUARTILE_FREQ(R1, R2, q, b) = the qth quartile of the data in the frequency table where q = 1, 2, 3 or 4
IQR_FREQ(R1, R2, b) = inter-quartile range (IQR) of the data in the frequency table
If b = TRUE, then the exclusive version of percentiles is used; otherwise (default) the inclusive version is used.
These functions are also used for calculating the Median, IQR and MAD in the Frequency Table Descriptive Statistics data analysis tool.
Half-Normal Distribution
Added support for the half-normal distribution via the HNORM_DIST and HNORM_INV functions.
Numerical Differentiation
The following worksheet function has been added.
DERIV(x, R1, h, R2) = the derivative f ′(x) where R1 is a cell that contains a formula that represents the function f(z) and z is the value contained in cell R2. If R2 is omitted, then it defaults to the first cell referenced in R1. h is the increment value; default .000001.
Numerical Integration
The following worksheet function has been added.
INTEGRAL(R1, lower, upper, iter, ttype, R2) = the integral ∫ f(x) dx between lower and upper where R1 is a cell that contains a formula that represents the function f(x) and x is the value contained in cell R2. If lower is omitted then -infinity is used, while if upper is omitted then +infinity is used. If R2 is omitted, then it defaults to the first cell referenced in R1. ttype is the estimation type (0 = Simpson’s rule, 1 = midpoint rule). iter = the number of subintervals (default 10,000).
Bug Fixes and Minor Enhancements
- Fixed some errors in the ORDER_SIM, RANGE_SIM and RANGE_DIST worksheet functions which were introduced in Rel 7.8 Also, these functions have been enhanced, especially to support discrete distributions.
- Added a default for the df argument in the QDIST, QPROB and QCRIT functions. The default represents infinity or at least a large value.
- Modified the Nemenyi post-doc option to the Friedman Test data analysis tool. The revised version now uses the df = infinity default in the QDIST and QCRIT formulas employed in the output from this data analysis tool.
- Changed the default for the last argument in the CountRowsUnique function from 1 to 0. This default was inadvertently changed in Rel 7.8 and has caused problems when using certain other capabilities (e.g. the Logistic and Probit Regression data analysis tool). This has now been resolved.
Doctor God bless you, would it be possible to include the vector support machine model with the minimum sequential optimization algorithm? It would be very useful.
Hello Antony,
Thank you for your suggestion. I will look into this.
Charles
precision is in decimals
https://prnt.sc/1sj5f2h
Hi Antony,
Does this come from Factor Analysis?
Charles
Yes doctor is from this web page.
Doctor God bless you, can you please check the varimax algorithm, as the program does not give accurate results. Saludos
Antony,
Can you send me an example where VARIMAX doesn’t give accurate results? How do you know that the VARIMAX algorithm doesn’t give accurate results (e.g. the results are different from those produced by a widely used platform).
I changed the VARIMAX procedure a year or so ago since its results didn’t match the version of VARIMAX on some other platform. Perhaps this means that there is more than one way to perform VARIMAX.
Charles