Home |
Search |
Today's Posts |
#19
![]()
Posted to microsoft.public.excel.misc
|
|||
|
|||
![]()
That's good to hear. The UDF should be much more convenient and 'sure' to
use since it will do its calculations in real-time without you having to run a macro to update values. "LenS" wrote: JL, I tried Alt F11 on your down load and got the VB Editor and copied the function to my Spreadsheet and it WORKED!!! Thanks again LenS JLatham (removethis) wrote: One thing more to try: Rename the copy of the file you previously downloaded then use the same link to download it again. I have changed the way I saved it - it was previously saved as Excel 2003 format only, I resaved and uploaded it in compatible with Excel 97...Excel 2003. Perhaps that is the problem with the Gray macro button - version conflict? "LenS" wrote: Happy New Year, The Function and Macro codes worked exactly like I wanted in the spreadsheet I down-loaded from your link. I was unable to get the code into my Excel though. The macro option was blanked out on your toolbar. I don't understand why this is? Can you help? If I can get this and duplicate were home free. Thanks, Len LenS wrote: Hello JL, Thanks for the code and your time. I have a question about UDF. 1) Can I step into a UDF and watch the process unfold in the code? I can not figure out why I keep getting Value# or Name# when I run the UDF. Here's the code below I entered into VB: Function ComputeI3(mylocation As Range) As Single Application.Volatile Dim LC As Integer If IsEmpty(mylocation.Offset(0, 1)) Then Exit Function ElseIf mylocation.Offset(0, 1) < 1 Then Exit Function End If 'initialize ComputeI3 = 0 ActiveCell.Offset(0, -1) = 0 For LC = 0 To mylocation.Offset(0, 1) - 1 ComputeI3 = ComputeI3 + mylocation.Offset(-LC, -1) Next End Function Thanks, LenS JLatham wrote: The code works perfectly - except there seems to be a bug in it! Kind of. If you have previously calculated the I3 values using it, then add more data down the sheet and run the macro to get the I3 values for the new data, then the previously calculated values become incorrect because they don't start out as zero, they start at whatever they were calculated to be the first time out. So, if you run it once, you get .17 for an I3 value, run it again, that goes to .34, run it again and that goes to .51, etc. This can be fixed easily by putting this line of code just ahead of the For LC = instruction: ActiveCell=0 to clear out any previous results. With that in mind, I've rewritten it as a User Defined Function (UDF). This means that you can refer to it in a cell just like a built-in Excel function. This method will do away with having to call the macro to calculate the I3 values and will calculate them on the fly. Replace the previous code with this code: Function ComputeI3(myLocation As Range) As Single Application.Volatile Dim LC As Integer If IsEmpty(myLocation.Offset(0, 1)) Then Exit Function ElseIf myLocation.Offset(0, 1) < 1 Then Exit Function End If 'initialize ComputeI3 = 0 For LC = 0 To myLocation.Offset(0, 1) - 1 ComputeI3 = ComputeI3 + myLocation.Offset(-LC, -1) Next End Function Then in your I3 entries in column AG, put formula like this, this example would go into AG2: =ComputeI3(AG2) You can then fill the formula on down the sheet with the AG2 automatically changing to AG3, AG4, etc as it is filled, and it will work for you. This also relieves you from having to have these calculated from columns AF, AG and AH - the function is always relative to the current cell no matter what column it is in, so it will use the value one column to the right as the PS/2 value, and always use the values from the column immediately to it's left to get the Q3 values from. The 'Application.Volatile' statement in it makes it update with any change made on the sheet, so if you change a PS/2 or even a Q3 value, I3 will be recalculated. If the needed PS/2 value has not been entered yet, or if a value of less than one (which would cause the loop to try to go from zero to a negative value, which it cannot do), then the value 0 is returned as the I3 value. "LenS" wrote: JL Thanks LenS JLatham wrote: That's quite true - one of the first things I was taught in a rather rigorous programming school was to "define the problem". Anyhow, the initial code will work as you want with one minor change: change For LC = 0 to Int(ActiveCell.Value/2)-1 to read For LC = 0 to ActiveCell.Value-1 and that will do it. I think you can probably remove the "I'm a little confused..." comment also, if you want to <g. I'll make similar change here and do a bit of testing, but I believe we're home now. If you don't hear back from me, you should be good to go. "LenS" wrote: JL, SP/2 is already divided by 2 and its integer taken and used in the count (your correct, SP/2 is a title, the 4 is the count value in search of x, 3 is the count value in search of x1 etc etc). As far as what values of Q3 are used I'm speculating as to what the Easy Language calls for in its For Count statement: For Count= 0 to Int(SP/2)-1 I think I need 0 to 4-1 =3; 0 to 3-1=2; 0 to 2-1=1 and 0 to 1-1=0. I think the author of this procedure intends for us to use all 4 positions or counts of Q3: Position (count) 0 where Q3=.06, Position (count) 1 where Q3=.04, Position (count) 2 where Q3=.05, Position (count) 3 where Q3=.02. Position 0 always refers to current days data, position 1 refers to the previous days data and so on. If I3=0 at the beginning of the procedure, then after 1 iteration I3 = I3 (0) + Q3 (.02) =.02. I use Q3 =.02 because (4-1 = 3 count or position), Q3 at position 3 = .02. The next iteration I3 = I3 (.02) + Q3 (.05) =.07. I use Q3 =.05 because (3-1 = 2 count or position), Q3 at position 2 =.05. The third iteration I3 = I3 (.07) + Q3 (.04)=.11. I use Q3=.04 because (2-1 =1 count or position), Q3 at position 1 =.04 The last iteration for this cell (x) is I3 = I3 (.11) + Q3 (.06) =.17. I use Q3 =.06 because (1-1=0 count or position), Q3 at position 0=.06. When For Count ends, we have I3 =.17 at the X position in column AG. I would like the next cell down calculated for I3 and so on until the whole data base is calculated for I3. I hope that clears the ambiguity. Let me know if I can clarify further. Thanks for making me think more about what I want!! It seems to be an essential prerequisite to programming. Thanks again for your time, LenS LenS wrote: JL I'm sorry for the confusion, I should take more time to detail the problem. As for the value in AH, in my original communication: There are no blanks in this column and how this value is used is as a counting value (as per the Easy language code : For Count= 0 to Int(SP/2)-1). If AH or SP/2 is 4, then it steps through and adds the Q3 value to I3 which is initially =0. I think that's what the Easy Lang. code is saying but I'm not sure. When this counter SP/2 is done, 4 in this part, I will have obtained I3 for the current cell in AG. I then want the routine to drop to the next cell in AG and begin the process again with a new counter (SP/2 is 3 in the next case) and calculate I3 for that cell. So column AH determines how many Q3's are involved at any one run. I apologize again for not being clear. I'm sure you run into alot of that programmer vs non-programmer. I really appreciate the time you put into this, thanks. Keep in mind that column AF and AH have no blanks except at the beginning and end of the data. LenS JLatham wrote: I'll try to help some, but I must admit to being confused by how you want the value in AH to be used to figure out which values of Q3 (In col AF) to add together to get the value for I3 in column AG. So the code may not be doing that properly. First the code, then some explanation. Sub CalculateI3() Dim I3value As Single Dim LastRowOfData as Long Dim LC as Integer ' Loop Counter 'find last row with data in column AH LastRowOfData = Range("AH" & Rows.Count).End(xlUp).Row 'go to first possible data entry in AH 'assumes row 1 has header text Range("AH2").Select 'work down thru all cells to last row used Do While ActiveCell.Row <= LastRowOfData 'assumes if cell in AH is not empty, it is number If Not(IsEmpty(ActiveCell)) Then 'I'm a little confused here, so may not be right For LC = 0 to Int(ActiveCell.Value/2)-1 ActiveCell.Offset(0,-1) = _ ActiveCell.Offset(0, -1) + _ ActiveCell.Offset(-LC,-2) Next ' end of LC loop End If ' test for empty cells 'move to next Row ActiveCell.Offset(1, 0).Activate Loop ' down thru rows End Sub The 'Dim' statements declare a variable for use later, simply reserving room for it to be used. If your code module includes the declaration Option Explicit at the beginning of it, then Dim or Const declarations are required. I recommend it because of several reasons that I won't go into here (end result: better code less prone to failure). Because we are going to be doing the math based on the contents of cells, we don't need variables to hold interim results. But we do need to know how far down the sheet to work at getting information and we need to know how many values of Q3 to add together to get the I3 values. LastRowOfData is set up as type Long so it can hold very large integer numbers, since you may have thousands of rows of data to work through. LC, to be used as a loop counter, is set up as Integer to hold smaller integer numbers, although to be on the safe side it possibly should be set up as type Long also. The first executable line of code looks up from the bottom of column AH until it finds some entry in that column. That tells us how far down the sheet we have to work in examining the information to be used. We save that row number for reference in LastRowOfData. We get into a Do While loop that is simply going to work down the SP2 column row by row looking for numbers until it gets past the LastRowOfData on the sheet. the If Not(IsEmpty()) statement is used to test if there's something in a cell in the SP2 column (AH) and if there is, an assumption that it is a number is made and we then calculate the I3 value. A little about the ActiveCell.Offset() instructions. Offset() takes two arguments, first the number of rows to offset, where negative numbers are "up" the sheet and positive numbers are on "down" the sheet, and zero is same row the active cell is on. Second argument is the number of columns to offset from the location; negative numbers are to the left of the reference cell, positive numbers are to the right of it. Zero is in same column. So an .Offset(0, 0) is actually no offset at all, and would refer to the reference location itself. |
Thread Tools | Search this Thread |
Display Modes | |
|
|
![]() |
||||
Thread | Forum | |||
Cell References | Excel Discussion (Misc queries) | |||
Using an offset formula for the reference in a relative reference | Excel Worksheet Functions | |||
Count # of A's in a Cell | Excel Worksheet Functions | |||
Cell Count | New Users to Excel | |||
Count on cell for each time it is changed | Excel Discussion (Misc queries) |