I need to find the First value and the last value of a range. A value is put into cell C12: number 1 for example and what happens is that excel will go down range H2:H9 and find the First Depth - "0" in this case and the last Depth 100 ......
I have a workbook with three sheets containing patient information from a clinic. Each sheet pertains to a different year. The sheets are formatted such that every row of each sheet contains a patient ID number in column A; in the same row, following the patient ID, is data pertaining to that patient, part of which stays the same (intake data)[=values in columns B:O] and part of which changes (encounter data)[=columns P:AA].
My challenge in cleaning and merging this dataset is that the ID numbers may not be trustworthy: I need a way to validate that there are no patient records [=rows] such that a given row has the same set of identifying information (OpenRecord Date, Sex, Age, History, etc.) located in columns B through O but a different ID number (column A) in a different sheet. In other words, I need to make sure the same patient isn't masquerading by a different ID number across different years [=worksheets]. What I am trying to do is develop a macro or formula that compares each row's range of values for B:O to every other row across each of the three worksheets; then find and list only matches for B:O that do not have the same value in Col A.
Complicating this operation are two considerations: 1) Each sheet in the workbook (excerpt attached) contains about 40,000 rows. 2) DOB is missing; instead, we have "Age." For those patients which appear in all three years captured in the three sheets, the "Age" value (column F) increases by +1 year.
A link to an example file on Google Drive : [URL] ....
This is probably easier than I'm making it on myself. I would like a formula that would find the value in D2 of the current sheet on a sheet named 6-19 in D:D, then take the value from L:L in that row and subtract it from L2 of the current sheet.
I've attached an example workbook : Cost Calc Upload.xlsx
I have 4 grids on the trending tab. I want to find the top 10 highest Color assets from Fleet Volume -Color tab under the correct month.
So for example, if the month is May as in cell G3 on the trending tab, then look at the data under the month of May on the Fleet Volume-Color tab and find the top 10 assets and drop in the City, Address, Model, Serial Number and then volume to the 1st grid on the trending tab, then repeat for Highest B&W
I want to the do the same for each of the other 3 grids on this tab. I want this to update based on the month on both tabs.
Of course, the data on the Fleet Volume tabs is a small range due to size, the data is a lot larger.
Is there are way to do this with a formula? I tried Large and small formula but not too sure how to bring over the other data like City, Model etc.
I have a simple problem I am trying to find a set a values on a column and what to replace once found with the first 5 most left characters. Example:
in column A I can have text and numbers; the numbers that I want to find are in the following format xy:zw / xy:zw on each case the values can be from 0 to 9 so this is tons of posible combinations, I been trying to use **:** ? **:** but I don't get none. This is what I am doing
Set target = Sheets("Sheet1"). Range("A1:A9000") For Each cell In target If cell.Value = "**:** ? **:**" Then cell.Value = "00.00" Next cell
A example would be 02:47 / 03:15 the next row could have 03:15 / 04:09 and so on however I could have text in some rows. In reality I would like to find and replace the cell.value with the Left 5 characters if the cell contains ANY value maching that format. Text in that column will never have that format combination
I am trying to fill a table of the last 12 values for the purposes of creating dynamic charts. I remember last time i used named ranges, offsets etc etc but been too long to remember how.
Ive attached a worksheet to explain it better.
I should probably mention, I want to be able to change cells C1 and C2 to update the values. Everything else wil be rather static.
There are three sheets in the workbook, Project, Tasks and Details and the expected resulting sheets are RESULT, In_Tasks_but_NOT_in_Projects and In_Details_but_NOT_in_Projects .
But now what I am looking for:
1. Copy the Projects data as is in the RESULT sheet.
2. Then in the Tasks sheet, if the ID matches paste the matching rows under the data from Projects (as in the result sheet with Orange colour)
3. If the ID is present in Tasks but NOT in Projects then copy it into the In_Tasks_but_NOT_in_Projects sheet.
4. Then If the ID and the Name in the Details tab matches with the data in the RESULT sheet then paste it under the ID and Name (as in the result sheet with Green colour)
5. If the ID does not match the ID in the results sheet then copy that row into the In_Details_but_NOT_in_Projects sheet.
The result of the current macro that RHCPgergo worked with are in the last sheet.
The formatting and colour of the rows doesn't matter, it is more of nice to have.
I am trying to find some missing values compared to 6 base values. For instance, I have a sheet with some names translated to another language, I am trying to find the languages some names have not been translated too.
For example, if I have six languages, Arabic, Japanese, Russian, Chinese-Simplified, Chinese- Traditional, and Korean to compare too, I want to find any names that are not translated in certain languages.
Sample:
John Japanese John Chinese - Simplified John Korean Martin Arabic Martin Chinese - Simplified Martin Russian Ramon Arabic Ramon Russian Sam Arabic Sam Chinese- Traditional
The idea is to add compare the cells of the first column with the third column. Where same letters/words exist, the corresponding value of the first column should be added to the second column (where no letter exists equally, the space remains empty), so it will look like this
A111 AA1122 B222 C333 D444
the third column always will have at least the same letters as the first column, but new letters/entries can occur.
Trying to find the sum of all cells in the array described in the formula that are equal to the values inside the quotations. I used this exact (as far as I can tell) formula to find the sum of values that were NOT equal to my quoted values and it worked just fine. Any ideas why formula 'A' will not work but formula 'B' does work? I have a feeling I'm missing something simple here!
Formula A - Does not work: =SUMPRODUCT(--('Master Lead Sheet'!$J$2:$J$10000=$B2),--('Master Lead Sheet'!$N$2:$N$10000="REJECTED"),--('Master Lead Sheet'!$N$2:$N$10000="CONDITIONED"),--('Master Lead Sheet'!$N$2:$N$10000="APPROVED"))
Formla B - Works: =SUMPRODUCT(--('Master Lead Sheet'!$J$2:$J$10000=$B2),--('Master Lead Sheet'!$N$2:$N$10000"No Answer"),--('Master Lead Sheet'!$N$2:$N$10000"Disconnected"),--('Master Lead Sheet'!$N$2:$N$10000"Wrong Number"),--('Master Lead Sheet'!$N$2:$N$10000"EMAILED"),--('Master Lead Sheet'!$N$2:$N$10000"needs to be emailed"),--('Master Lead Sheet'!$N$2:$N$10000"Refund"),--('Master Lead Sheet'!$N$2:$N$10000"REFUNDED"))
How do you write an excel macro that looks at the number in the first column (day #) and finds all the duplicate id#s in the second column that are in day 1and adds the amounts together in the 3rd column then writes the first column number (day#), second column number(id#) and the third column (sum of the amounts of duplicate Id#) to an new worksheet. Then the macro would loop through day #2 and do the same thing. Notice that the values in the id column are unique in this data set below this is how I would like the data to look. I have accomplished this in a pivot table but my problem is I need a cvs file to export the final data into an external database which is why I need a macro.....
What I have In Column B, I have the datesIn Column I, I have engineers name What I need I want a macro to generate Serial Nos. (1,2,3....... n) in column A If an only if the date in column B is today's date and the engineer's name matches with the PC's username
I'm running into an issue trying to calculate unique values in a Data column based on a few variables in other columns.
My current formula in Summary tab D4:D19 is {=SUM(IF(FREQUENCY(IF(Data!$I$3:$I$66<$E$1,IF(Data!$A$3:$A$66=$H$1, IF(Data!$C$3:$C$66=A4,ROW(Data!$I$3:$I$66)))),ROW(Data!$I$3:$I$66)),1)) + SUM(IF(FREQUENCY(IF(Data!$I$3:$I$66<$E$1,IF(Data!$A$3:$A$66=$I$1, IF(Data!$C$3:$C$66=A4,ROW(Data!$I$3:$I$66)))),ROW(Data!$I$3:$I$66)),1))}
This is currently counting the number of times a date value (data column I) appears for that name (A4:A19) in the data when meeting all of the conditions. I need it to instead count the number of times a unique date appears for that name with the additional conditions met (which all appear to work fine).
The results in the pink highlighted cells (Summary column D) should be:
Names starting with A - 3 All others - 2
I've left some other columns in the data with X's so that I can easily convert this back to my working spreadsheet.
Is it possible to have a formula where it will retrieve a specific value on one sheet based on matching values? Unfortunately, hlookup wont work as there are several values in the table where the 'lookup_value' is the same.
Example:
On sheet 1 (titled BSRC), I have a table which has a list of values in column 1. On column two I have the 'lookup value' which needs to be matched and all values underneath it to be retrieved from sheet 2 (titled CSRD). The 'lookup_value' range is on row 9 of sheet 2.
I have attached a sample spreadsheet for clarification.
I want to build a database where the user can select a specific time period. I have a static file with the time periods and the related values. For instance I have in column A1:A15 the time periods from January 2011 to Dezember 2011, but I have some dates multiple times as there are more values attached to it. For instance I have in the first 4 rows January 2011...what I want is when someone puts in the value January 2011 in cell B1 and Dezember 2011 in cell B2 that the whole static file gets copied to another location (including the multiple dates) displaying the chosen time period. Similarly if someone puts in the value March 2011 to November 2011, I want only those values to be copied.
is it possible to write a macro that will convert absolute values into negative based on criteria? Data gets downloaded from in house system which comes out as all absolute values, now i need the macro to covert absolute values to negative if the amount is either "Our Reciepts" or "Our Delivery"
Below is a small sample of data, real data is around 100 rows
Col DCol EOur Receipts1231Our Delivery1231Delivery Settlement1323
In sheet1, I have the borrowers and outstanding per employee. In the Area Avg sheet, I have the average borrowers and outstanding in each area as this is the standard for comparison (I got this through the subtotal function).
Now what I have to do is in sheet1, format(change the color) each cell under borrowers which is greater than the average of the relevant area. Since, (in my actual data) the name of the area may be duplicate in different regions, the reference value (average borrowers of the area) from "Area Avg" sheet has to be traced through multiple criteria which are: Division, Region and Area. The same also has to be done for outstanding but the solution can be provided for borrowers only.
The simplified example below is intended to calculate the sums from the data values in the upper data set and place them in the appropriate cells in the lower summary table. Example......
I would like to find a single-cell formula that will handle the multiple instances of either the X or Y headings (ex: there are two 'Yes' Rows) and still reach the appropriate summations. I would like to accomplish this without resorting to Macros or manual cell references if possible. I've acheived some limited success with array formulae so far, but find that the multiple instances of 'Yes' cause only one row to be summed before the formula finishes.
I have different reports, some have fifty transactions, others have thousands. My goal is to: Insert a new row every time the values in the "Account" column meet a certain criteria, AND THEN add the totals for the Debit and Credit Columns.
Let's say I start with a table that looks like this:
A B C D Dept E Account T F F2 G Debit Credit Total
33010
[Code] ....
I want to group the first four rows because Accounts 33010 and 33015 are in the same department. Same with 50050 and 500060. I want to then insert a row below the last row with "33015" as its Account #. And add the values for Debit and Credit. It'd look like this:
A B C D Dept E Account
[Code] .........
Honestly, I have tried everything. Running a Macros with Relative reference does not cut it.
I'm starting a dashboard, where on the front page I have two combo boxes on the left, and three empty fields to the right. I'd like the three fields to the right to auto-populate table-based values depending on the chosen criteria from BOTH fields (by store and month/date). I've attached a sample of what I've got so far. I've only provided three tables for this example, and I have a table with the same column/row titles for each metric and I have three different metrics I'd like to auto populate: COGs, Sales, and GM% or in the example, metric 1, metric 2, metric 3. No pattern in the table values, just wanted to populate the fields quickly. All fields are organized by store/month-date and I've set up a link to my combo boxes on a calculations tab.
I have x coordinates in column 1 with coresponding y coordinates in column 2. From my data of x,y coordinates I want to find points with similar y values. In my data I have defined groups of numbers, i.e. I have a set of numbers with y values around 30 (+-10), then a new group with y values around 60 (+-10), and so on... Sometimes the groups are not totaly separeted, there could be a few points with y values between 30 and 60. These points can be grouped with the closest group of coordinates (30 or 60). Each group of x,y coordinates could be copypasted in the columns to the right (columns 3 and up).
So what I want to do is find a group of coordinates. This group will have at least 40 individual points +-10 from a group mean. The coordinates need to be sorted out from the data and put in seperate columns.
I'm attempting to assign 1 of 4 predetermined values to a cell based on the values of entries in another cell.
View the example attached : Value assignmts.xls
Assign the Value of 1, to (D8) when any value entered in (C9) is equal to or over 380, but less than 410. Assign the Value of 2, to (D8) when any value entered in (C9) is equal to or over 410, but less than 440. Assign the Value of 3, to (D8) when any value entered in (C9) is equal to or over 440, but less than 470. Assign the Value of 0, to (D8) when any value entered in (C9) less than 380.
I want to sum the values in cells E2:P110 based on the values column D. The values in D are formulas resulting in something that appears to match D112 in some cases. I'm using the following equation:
=SUMIF(D2:D110,D112,E2:P110)
My problem is that D2 :D10 have a formula in it and it's not matching. If I enter the result of the formula, all is good. How should I deal with this?
I have a table of times which shows how long it took someone to process a form. If this value is longer than 10 minutes the cell next to the duration time contains the value 'Break Over Ten'. This is only true when the time is outside of 12:00 to 14:00. If the break is within this time period it shows 'Lunch'.
Is it possible to sum all of the time values that are over flagged as 'Break Over Ten'? This is how it looks
Event Date Start Time End Time Duration Breaks Duration Of Break 19/12/201308:43 08:43 00:00:35 00:00 19/12/201308:43 08:44 00:01:03 00:01 19/12/201309:05 09:19 00:13:15 BREAK OVER TEN00:13 19/12/201309:05 09:19 00:13:15 BREAK OVER TEN00:13
I was thinking of using SUMIF but can't work out how to total the values in the duration of break column when the cell to the left reads break over ten. Is it possible to do this?
Essentially, the actual amount contains a set of formula, that sums up the spending sheet (the one on top) based the category and sub-category on the budget sheet (the one on the bottom). Tried to use SUMIF but it seems to work on a single column.
Col A( Data) Col B ( Invoice Description) Col C( Result)
I have date and Invoice description column filled with data, I need to filter Column B> custom filter> contains > "Dup" and enter value as "Twice" in Column C. again i customer filter> contains > " Canteen" then enter value in Column C as "Tea" similarly i have 45 conditions and the number of rows are not fixed, I would appreciate very much if someone could help me to build a code with case function i can add all those conditions: ....
I am looking for a smart way to extract the values corresponding to the first and last day of a month I specify from my time series. Column B has the dates (dd/mm/yy), column A is text(B,("mmm-yy") and column E has the values I need to extract. I have a column with all the months of the year: Dec 12, Jan 13, Feb 13 etc... and I need to pick the value in E that corresponds to the last observable day of each month and divide that by the value corresponding to the first observable day of the month.