tag:blogger.com,1999:blog-3721871707048712457.post8869964030748654874..comments2024-03-28T00:39:48.395-04:00Comments on Breaking BI: Performing Simple Linear Regression in TableauBreaking BIhttp://www.blogger.com/profile/02551920506874509998noreply@blogger.comBlogger20125tag:blogger.com,1999:blog-3721871707048712457.post-36316565170694392312016-05-04T07:06:04.007-04:002016-05-04T07:06:04.007-04:00Hi, I was wondering if you were able to figure thi...Hi, I was wondering if you were able to figure this out. Greatly interested in understand how you did this.<br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-57342694857012409652015-11-19T14:32:34.406-05:002015-11-19T14:32:34.406-05:00Brad,
How would you handle X if it is a date rang...Brad, <br />How would you handle X if it is a date range from September 2013 to August 2015?<br /><br />Thank you!<br />JBfh2levelhttps://www.blogger.com/profile/00072179615984868769noreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-89437909859702179062015-11-16T09:36:31.476-05:002015-11-16T09:36:31.476-05:00Hello Brad,
Thank you for this tutorial. I have ...Hello Brad, <br /><br />Thank you for this tutorial. I have an issue; I want to build a linear regression in Tableau following the steps decribed above but I couldn't succed. Can you help me please? I mean I added a trend line then I calculated all the fields but my trend line didn't change. Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-28317858021810046062015-10-28T20:24:35.474-04:002015-10-28T20:24:35.474-04:00Any way to get regression statistics to show corre...Any way to get regression statistics to show correlation coefficient, etc?Mehttps://www.blogger.com/profile/01399160310118574639noreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-22243904612893140302015-09-03T05:31:37.128-04:002015-09-03T05:31:37.128-04:00Hi Brad,
I've followed the instructions and i...Hi Brad,<br /><br />I've followed the instructions and it works great, except for when I have null values for the dependent variable Y. How can the calculated fields be adapted such that when a data point does not have both an X and Y value it is removed from the calculations?<br /><br />Thanks,<br /><br />Dipesh<br /><br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-91588034905695000752015-07-21T12:32:51.409-04:002015-07-21T12:32:51.409-04:00Thank you very much , works like a charm ?
* How...Thank you very much , works like a charm ? <br /><br />* How can we export the SLOPE as a Field for further uses ?<br />* How can we do computation at multiple levels ( for each City suppose i get this graph and i want to compare cities ) Anonymoushttps://www.blogger.com/profile/04464960465656114829noreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-68303017802790640332015-07-09T18:54:27.468-04:002015-07-09T18:54:27.468-04:00Could you please explain why -[Slope] * [X mean] ...Could you please explain why -[Slope] * [X mean] + [Y mean]?<br />Why is Minus used?<br />Though the value is correct, why is it not y = mx+bAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-37933955673202446142015-06-04T01:53:37.600-04:002015-06-04T01:53:37.600-04:00hi again, i solved it myself, no need to reply :D ...hi again, i solved it myself, no need to reply :D By the way this is really a great blogniahttps://www.blogger.com/profile/10202633937791251852noreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-80798322857709778222015-06-04T00:55:50.768-04:002015-06-04T00:55:50.768-04:00Hi, great blog *bookmarked*, anyway i have a quest...Hi, great blog *bookmarked*, anyway i have a question. What if there is null in my Y variable, what will this formula do with the null data? i compare the slope and intercept using this formula with leaving the null data (not using it in modeling) just like what tableau trend line did, and the result is different, and i compare it again using excel regression with Y null replaced by 0, and the result still different.<br />(For example : i use all data with this formula and the result -> slope = 2,13 and intercept = 2063, leaving null data i get the same as tableau trend line slope =2,17 and intercept =2024, i replaced null with 0 i get slope =1,95 and intercept=2239) how is this happened? please help. Thank you so muchniahttps://www.blogger.com/profile/10202633937791251852noreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-67402546264432056082015-04-22T10:35:39.726-04:002015-04-22T10:35:39.726-04:00Can you answer this question in the forum :
http:/...Can you answer this question in the forum :<br />http://community.tableau.com/thread/163382Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-12248425036040910062014-10-02T10:39:21.054-04:002014-10-02T10:39:21.054-04:00I'm not sure I understand your question. The ...I'm not sure I understand your question. The Predictive Analytics Part 1 post walks through creating the model. If you want the line from that, you can simply use the fit$fitted function to return the fitted values. Does this make sense?Breaking BIhttps://www.blogger.com/profile/02551920506874509998noreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-57235185478214822392014-10-02T10:13:32.317-04:002014-10-02T10:13:32.317-04:00Thanks for the post. I am a newbie to R and Tablea...Thanks for the post. I am a newbie to R and Tableau. Is it possible to recreate the simple linear regression using R? I saw your post on the 1st part of predictive analytics, but is it possible to just plot the simple regression line produced by R? Thanks.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-12220176982724616122014-07-23T10:38:51.821-04:002014-07-23T10:38:51.821-04:00Excellent work! Thanks Brad for bringing the regre...Excellent work! Thanks Brad for bringing the regression line equation derivation into Tableau!Ashishhttps://www.blogger.com/profile/06841478924679921671noreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-34863160134826874162014-03-20T15:26:59.966-04:002014-03-20T15:26:59.966-04:00Actually, that way is much easier. As a mathemati...Actually, that way is much easier. As a mathematical fact, a regression line must go through the ( average(x), average(y) ). So, if you force the y-intercept to be 0, then your line also goes through (0, 0). If you remember your eighth grade math, two points make a line. So, all you need to do is create an estimated y calculation of<br /><br />average( [x] ) / average( [y] )<br /><br />Then add the line (x, estimated y) to your graph. Let me know if you need an example.<br /><br />Thanks!Breaking BIhttps://www.blogger.com/profile/02551920506874509998noreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-21574373453682478682014-03-18T20:04:50.733-04:002014-03-18T20:04:50.733-04:00Is there a way to do this with the y-intercept for...Is there a way to do this with the y-intercept forced to zero?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-72110752524929664452014-02-03T17:23:54.210-05:002014-02-03T17:23:54.210-05:00Thanks for commenting! The likely culprit is Comp...Thanks for commenting! The likely culprit is Compute Using. Note that I compute using Product Name. If you're unaware how Compute Using works, you can check out<br /><br />http://breaking-bi.blogspot.com/2013/07/introduction-to-table-calculations.htmlBreaking BIhttps://www.blogger.com/profile/02551920506874509998noreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-75247874091316200582014-02-03T17:08:23.567-05:002014-02-03T17:08:23.567-05:00In version 8.1 of Tableau, the X and Y Differences...In version 8.1 of Tableau, the X and Y Differences return 0. I deleted Workbook, and recreated a 2nd time. Same issue...? <br /><br />X Mean: WINDOW_AVG(SUM([Sales]))<br />Y Mean: WINDOW_AVG(SUM([Profit]))<br /><br />X Differences: SUM([Sales]) - [X Mean]<br />Y Differences: SUM([Profit]) - [Y Mean]<br /><br />Sample - Superstore SubSet<br /><br />columns: SUM(Sales)<br />rows: SUM(Profit)<br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-52896222594630987312013-09-24T08:51:44.995-04:002013-09-24T08:51:44.995-04:00Hi Shawn,
I think you meant to reply to Tableau v...Hi Shawn,<br /><br />I think you meant to reply to Tableau vs. Power Pivot: Part 2. A clustered bar graph is a bar graph that displays multiple dimensions on the same axis. It's somewhat difficult to explain. If you look at the examples for Excel and Power View, you will see clustered bar graphs. Thanks!Breaking BIhttps://www.blogger.com/profile/02551920506874509998noreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-91012232889723730782013-09-24T08:22:11.102-04:002013-09-24T08:22:11.102-04:00Brad what do you mean by "clustered bar graph...Brad what do you mean by "clustered bar graphs"?Shawn Wallworkhttps://www.blogger.com/profile/07875560767627938393noreply@blogger.comtag:blogger.com,1999:blog-3721871707048712457.post-6323157485036169152013-07-10T12:48:51.038-04:002013-07-10T12:48:51.038-04:00Very well done.Very well done.Anonymousnoreply@blogger.com