To comment on the following update, log in, then open the issue:
http://www.openoffice.org/issues/show_bug.cgi?id=76142
                 Issue #|76142
                 Summary|LINEST Example Fails
               Component|Spreadsheet
                 Version|OOo 2.2
                Platform|PC
                     URL|
              OS/Version|Windows XP
                  Status|UNCONFIRMED
       Status whiteboard|
                Keywords|
              Resolution|
              Issue type|DEFECT
                Priority|P3
            Subcomponent|code
             Assigned to|spreadsheet
             Reported by|pperson





------- Additional comments from [EMAIL PROTECTED] Fri Apr  6 01:03:17 +0000 
2007 -------
Per your help file, this is what LINEST looks like:

x1 x2 y   LINEST value (blank cell) (blank cell)
4  7  100 4,17         -3,48        82,33
5  9  105 5,46         10,96        9,35
6 11  104 0,87         5,06         #NA
7 12  108 13,21        4            #NA
8 15  111 675,45       102,26       #NA
9 17  120 10           19           133

Per reality, using the same settings as in the Help file produces:

x1      x2      y       -27.96  66.81   0
4       7       100     #N/A    #N/A    #N/A
5       9       105     #N/A    #N/A    #N/A
6       11      104     #N/A    #N/A    #N/A
7       12      108     #N/A    #N/A    #N/A
8       15      111     #N/A    #N/A    #N/A
9       17      120                     
10      19      133                     

I am interested in this because I have three datasets I have brought over from 
Quattro Pro and I trying to reproduce Quattro Pro's ability to do regressions. 
Since I only have one X column, I am getting not only dubious slope values but 
also #N/A for the Y-intercept. I realize that you are satisfied to do no better 
than Microsoft's spreadsheet, but, really, should you not consider doing better 
than Microsoft? 

Notes: 
I tried FORECAST, but it does not produce the right results either; perhaps it, 
also, is under the misapprehension that all lines must go through the origin.)

No combination of, for example, "4" and "7" and -27.96 and 66.81 can possibly 
produce a result of 100 if the y-intercept is 0,0.

I know of no reason that either at least two or exactly two data series are 
needed, together with one control series, to produce the results of a 
regression. This should work with one column, or the Help should clearly 
identify the one-column version.

---------------------------------------------------------------------
Please do not reply to this automatically generated notification from
Issue Tracker. Please log onto the website and enter your comments.
http://qa.openoffice.org/issue_handling/project_issues.html#notification

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to