
1. Re: relational statistics
guest contributor Apr 26, 2011 9:39 AM (in response to Tim Harris)I have been looking for the same, but other than simple regression, it does not provide any other statistics. You can work out complex SQL calculations for the same probably but nothing that is easy.

2. Re: relational statistics
Joe Mako Apr 26, 2011 10:46 AM (in response to Tim Harris)Another option is custom table calculations. I believe most calculations you are looking for can be calculated in Tableau with understanding of how those calculations are performed. The downside is, if you have over 100,000 rows, the table calculations may be slow to process the advanced calculations.
Another option that may work is to have your data source be a database that has those functions builtin, or the ability to add user defined functions, and use RAWSQL() passthrough functions to expose them to Tableau.

3. Re: relational statistics
Scott Tennican Nov 28, 2011 12:45 PM (in response to Tim Harris)Tableau does not yet have any builtin multivariate summary statistics or tests.
Other than Pearson's correlation coefficient and a chisquared test for independence of two categorical variables,
which statistics features would would you like to see in Tableau first?
You can compute the Pearson's correlation coefficient in a calculated field like this:
WINDOW_SUM(
(SUM([x])WINDOW_AVG(SUM([x]))) *
(SUM([y])WINDOW_AVG(SUM([y]))) )
/
( SQRT(WINDOW_SUM(SQUARE(SUM([x])WINDOW_AVG(SUM([x]))))) *
SQRT(WINDOW_SUM(SQUARE(SUM([y])WINDOW_AVG(SUM([y]))))) )

4. Re: relational statistics
brettsisman0 Dec 21, 2011 2:35 PM (in response to Tim Harris)I would really love to see multivariate analysis included. Any chance it's coming out with 7.0 or are we going to need to wait a while?

5. Re: relational statistics
Scott Tennican Dec 21, 2011 2:59 PM (in response to Tim Harris)Well, there are many kinds of multivariate analysis (e.g. regression analysis) in Tableau many of which have improved in 7.0.
Are you asking about simple multivariate summary statistics and hypothesis tests like the ones below?
This kind of thing currently requires table calcs like the one I pasted above.
Pearson's linear correlation coefficient
Spearman's rank correlation coefficient
ttest or a Welch’s test for difference in means or proportions
Chisq test for independence of two categorical variables
Chisq test for a specified proportion of values of a categorical variable
Chisq test for homogeneity of two samples of a categorical variable
Oneway ANOVA test of homogeneity of several means

6. Re: relational statistics
Hunain Kochra Nov 11, 2012 2:36 AM (in response to Scott Tennican)Hello Scott,
Thanks for this post. I have 2 variables(X & Y) in a dataset of 30,000+ records plotted on a scatter.
Using the trend line I can identify if the association is positive or negative between A & B, however I am unable to indicate the actual correlation value on the same scatter plot.
Is this possible? Thanks for any help.
Hunain

7. Re: relational statistics
Xiaodong Han Nov 28, 2012 7:55 AM (in response to Hunain Kochra)Scott:
the formula helps a lot. Would you mind post the formulas for the other statistics you mentioned?
Thanks,
Xiaodong

8. Re: relational statistics
elias.fayad Feb 28, 2013 10:47 AM (in response to Scott Tennican)Thanks for that, Scott. Would you say that the manual Pearson's linear correlation coefficient calculation for 2 variables would need to be applied manually across the board if, say, I had 10 variables whose correlation I needed to explore? Any thoughts on how I may be able to produce a crosscorrelation 10x10 matrix, in Tableau 7 or 8?