On the Nonadditivity of Correlation Coefficients Part 1: Pearson and Spearman Coefficients

Tags

, , , ,

This is Part 1 of a tutorial series on the nonadditivity of correlation coefficients. We demonstrate why it is not possible to arithmetically add, subtract, and average Pearson’s r or Spearman’s rs.

The article is available at

http://minerazzi.com/tutorials/nonadditivity-correlations-part-1.pdf

07-04-2017 update: In page 1 the line for the Beta1 should read “is the slope of a simple linear regression model”. My fault. Fixed today along with few other nuances.

Enjoy it!

t,p, & Effect Size Estimator

Tags

, , , ,

This is a new Minerazzi tool, available now at

http://www.minerazzi.com/tools/t-p-effect-size/estimator.php.

This tool estimates t-scores from p-values and vice versa for a given number of degrees of freedom υ. Just enter a (t,υ) or (p,υ) pair and this tool will solve for the missing term.

The tool also estimates the statistics that one would obtain if the computed estimates correspond to a set of paired variables (x, y), or to effect sizes from any two samples of same sizes (n1 and n2); i.e. samples with same degrees of freedom (υ1 and υ2).

The tool’s page lists some interesting exercises and good references for effect size conversions and meta-analysis in general. Enjoy it!

 

Student’s t Table Generator

Tags

, ,

This is a new tool available now at

http://www.minerazzi.com/tools/student-t-table/generator.php

This tool allows you to easily generate a customized table of Student’s t-values.

That is done by iteratively calling (i.e., looping) the very same algorithm that we use for our t-values Calculator.

This tool comes handy when you don’t have statistical t-tables around or are working with p and t values, or degrees of freedom, not available from such tables. Avoid pausing a problem for annoying linear interpolation workarounds!

Meta-Analysis Miner

Tags

, , , ,

This is a new miner available now at

http://www.minerazzi.com/meta-analysis/

Find resources relevant to meta-analysis, effect sizes, and power analysis with this Minerazzi miner.

Build curated collections on these subjects.

The miner includes Bing and Google RSS News channels powered by our SPP tool.

t-values Calculator for Student’s t Hypothesis Testing

Tags

, , , ,

The t-values Calculator is a new Minerazzi tool, available now at

http://www.minerazzi.com/tools/t-values/calculator.php

The tool works by computing, through numerical approximation methods, the Inverse Cumulative Distribution Function or ICDF (also known as the Quantile Function, QF) of the Student’s t Distribution.

This new tool complements our p-values Calculator, described in

https://irthoughts.wordpress.com/2017/06/12/p-values-calculator-for-students-t-hypothesis-testing/

and available at

http://www.minerazzi.com/tools/p-values/calculator.php

The t-values calculator comes handy when you don’t have around statistical t-tables or want to t-testing at confidence levels and degrees of freedom not listed in such tables. In fact, with a small change to the script, we can build another tool for the generation of t-tables at any confidence level and degrees of freedom as specified by a user.

Both tools will eventually be used to provide a third tool: The t-p Transformation tool that, as the name suggests, transforms CDF (t-to-p) into ICDF (p-to-t) results and vice versa.

 

 

p-values Calculator for Student’s t Hypothesis Testing

Tags

, , , , ,

The p-Values Calculator is a new Minerazzi tool that is available now at

http://www.minerazzi.com/tools/p-values/calculator.php

Submitting a Student’s t value and degrees of freedom returns a p-value. This is a great tool for Student’s t hypothesis testing.

The tool works by numerically approximating the CDF (Cumulative Distribution Function). This is the integral of the PDF (Probability Distribution Function) of the Student’s t Distribution. The theory behind these calculations along with valuable references are given in the tool’s page.

PS.

The reverse process, computing t from a p-value is possible by inverting the CDF to compute the Quantile Function (QF), also known as the inverse CDF. Our (soon to be released) t-p Transformations tool computes both the CDF (t-to-p) and QF (p-to-t).

Updating Several Miners

Tags

, , ,

We have reindexed the following miners:

http://www.minerazzi.com/webspam
http://www.minerazzi.com/zika
http://www.minerazzi.com/usc
http://www.minerazzi.com/seominer
http://www.minerazzi.com/religions
http://www.minerazzi.com/shopper
http://www.minerazzi.com/rblogs
http://www.minerazzi.com/quantum

 

 

Memory Based Reasoning in AI

Tags

, , , , , ,

This is a nice course on Memory Based Reasoning in AI taught by Dr. Deepak Khemani and Dr. Sutanu Chakraborti at the Department of Computer Science & Engineering, Indian Institute of Technology Madras, Chennai, India. See

http://aidblab.cse.iitm.ac.in/cs625/

Happy to see that the following readings from the course are based on two articles we published long time ago at Mi Islita.com, now a search engine site about Puerto Rico resources.

Reading #10
Vector Space Models
http://aidblab.cse.iitm.ac.in/cs625/10.VectorSpace-model.pdf

Reading #11
SVD-LSI
http://aidblab.cse.iitm.ac.in/cs625/11.SVD-LSI.pdf

Python Search Engine

Tags

, , , , , , ,

Here is a python-based search engine with an implementation inspired on one of our papers at the old Mi Islita.com site, now a search engine on Puerto Rico.

http://nullege.com/codes/show/src@g@e@gensim-HEAD@gensim@test@test_miislita.py

This module replicates the miislita vector spaces from
"A Linear Algebra Approach to the Vector Space Model -- A Fast Track Tutorial"
by Dr. E. Garcia...
 

Great and positive accomplishment!

That tutorial is no longer at miislita.com, but was long ago moved to minerazzi.com. Find it here:

http://www.minerazzi.com/tutorials/term-vector-linear-algebra.pdf

For other resources do a search for python in our IR miner at

http://www.minerazzi.com/irc

For inquiries about that implementation, contact its author.

For other inquiries, applications, suggestions, drop me a line.

PS. Please note that Nullege.com itself is a search engine for finding python code. Here is a good example: http://nullege.com/codes/search/wx.calendar.CalendarCtrl