Calculating Correlation Between Two Series Of Data
Dec 13, 2012
I have a list with hundreds of customers. Each customer has two measures (i won't bore you with how they are calculated)
1) % Coverage by Salesman
2) % Of Opportunity Won
What tips and tricks can you give me to take my analysis to the next level? The example is a very limited sample and the data was made up using a formula so correlated very well, i've chucked in a couple of anamlous results to make it a bit more realistic.
One of the main things i would like to see is a correlation "score". I suppose if the two series correlated perfectly then the score would be 100%. The less they correlate, then the lower this score gets....
In business terms, if i can run this rule over thousands of customers and get a good correlation - then it becomes a conversation with the salesman to show them that doing those pesky, boring coverage calls has a measurable impact on sales...and here is the number to prove it...
How I can interpret this data?
View 1 Replies
ADVERTISEMENT
Mar 3, 2014
I am now analysing the ordered correlations of products in a warehouse.For example, I have 16 products (product 1 to product 16) in my warehouse. During last month, there were totally 391 orders (order 1 to order 391). If a product is requested in a ordered, an "X" will be marked as per attached sheet.
Now my goal is to figure out which pair (or even triple) of products are ordered/requested at the same time (or in the same order) for the last month. I intend to use the analasys result to move the more ordered pair/triple products to stay near to each other at the warehouse so it will save the time for picking.
How to do that analysis?
View 8 Replies
View Related
Feb 20, 2008
I have an array of data (A,B,C,D,...,H) and a correlation/relationship data table (asymmetric data).
Out of the array data, i need to make a combination of 3 data out of 8 AND input the relationship data according to the combination.
I found a problem to take the reference from correlation data.
to help you understand my issue, i attached the file with this posting.
View 5 Replies
View Related
May 31, 2008
I just recently installed Excel 2007 and I would like to know if it's possible to change all data points of a chart at the same time. In Excel 2003, I would normally hold down shift while clicking on each of the data points to make a global change. However, it appears I cannot do that in 2007.
I would like to display each data point's series name. When I go to Layout on the Excel Ribbon, and click on "Data Labels", and click on "More Data Label Options", the actual Y-axis values are shown for each data point. However, I do not want this - I actually only want the Series Name, but when I uncheck "Value" and check "Series Name" instead (under "Label Contains"), it only changes it for one of the series. Is there a better way, instead of going through each and every single series to make this change?
View 4 Replies
View Related
Jan 9, 2013
i have 2 columns of number of cars using various roads, taken in the last 30 days. A macro has processed the data in these ranges and changed the outlier values to "Bad" so the formula ignores them.
However in the columns A and C , there are dates that the counts were taken on.
{=IFERROR(CORREL(IF(B4:B33"Bad";IF(D4:D33"Bad";B4:B33)));IF(B4:B33"Bad";IF(D4:D33"Bad";D4:D33)));"0")}
sometimes these dates aren't the same (no data available on christmas in some areas, boxing day in others etc. Obviously this destroys the correlation, as it puts the days out of synchronisation and the correlation falls apart.
Is there any way i can adapt the formula for it to check the dates are the same before correlating? It'd almost be like a vlookup, find the date, if match, add to calculation, if not then go to the next date. i'm happy for it to only correlate 26 of the 30 days if there are that many conflicts.
example for last 6 days performed on the 28th of december (looking for 6 calendar days up to but not including 28th dec.
22/12/2012 9 22/12/2012 Bad 23/12/2012 10 23/12/2012 22 24/12/2012 3 24/12/2012 21 25/12/2012 7 26/12/2012 18 26/12/2012 8 27/12/2012 15 27/12/2012 15 These 2 cells would be blank and this column would only contain values for 5 days.it could only correlate the 3 days of 23, 24 and 27.
Given that i'm going to be correlating this information with a number of locations, each with different reporting dates, filtering the dates before correlating and deleting unmatched dates is not a great option, but I will consider it if there is no other way.
View 9 Replies
View Related
Feb 25, 2008
concerning the use of correlation in vba. What are my options when i want to calculate a correlation within vba. The variables are all still within vba.
Who knows what formula's i could use to keep everything within vba.
If you don't understand my question please ask so i can try to make it more clear.
View 9 Replies
View Related
Aug 4, 2014
I run a large simulation experiment. I have a loop plotting data in excel of a user defined area. Because of the limit of 255 series I have allowed a maximum of 250 simulations (they all need to be plotted). But the length of each simulation is free. I know there is a limit of 32.000 data points in a graph and I have this as a condition too.
If I set the data range to 100 columns and 3000 rows the graph is produced when I plot by columns. (code below)
But if I set the data range to 250 columns and 1000 rows I get the above mentioned error message. Even though I only have 250 series.
After the data is plotted it is the code below that gets the error:
[Code] .....
View 1 Replies
View Related
Dec 22, 2011
I have the following code:
Sub Macro5()
ActiveSheet.ChartObjects("Chart 243").Activate
ActiveSheet.ChartObjects("Chart 243").Activate
ActiveChart.SeriesCollection(1).Delete
ActiveChart.SeriesCollection(2).Delete
End Sub
However, if there is no SereisCollection(1) actually present in the chart I get an error. How can I work around this? I will need an IF statement I assume, just not sure what it will look like.
View 4 Replies
View Related
Jan 16, 2006
I have code to create a correlation matrix (NxN, where N is the number of columns). This is done by selecting an area that is NxN, entering the function and range, then hitting ctrl +shift + enter (array formula).
However, I want to convert this to accept VBA arrays, rather than a data range, and give the output in form of an array as well.
VB:
Function CorrmatK(dataRange As Object) As Variant
On Error Goto 20
Dim r As Integer, n As Integer, rr As Integer, i As Integer, j As Integer, k As Integer, doit As Integer
Dim x() As Variant, mc() As Double, ss() As Double, m() As Double, ob As Object
r = dataRange.Rows.Count
n = dataRange.Columns.Count
[Code] ......
View 7 Replies
View Related
Oct 30, 2006
I am in the unfortunate position of having to run about 900 correlations with Excel as my only option. Normally this would be fine, but in this case, I need the Pearson's test of significance. The only way I know how to obtain this in Excel is through using a regression, which is incredibly time-consuming. Does anyone have a suggestion how to make the best use of time?
View 11 Replies
View Related
Mar 11, 2014
I am using excel to do a monte carlo analysis of a process. Until now, all my distributions have been uncorrelated, but from now I would like to produce distributions with a given correlation. This is a fairly easy process for normal distribution, but in this case one of the distributions is Weibull and the other lognormal. I don't have a stats background, so how to implement this. Producing the two distributions is no problem, but I would now like to be able to alter them or the original formulae so they are correlated properly. If it would greatly simplify the process, the lognormal distribution could be replaced by a normal, but not the weibull.
The formulae I'm using for each is:
View 4 Replies
View Related
Jul 25, 2006
I wasn't able to attach the file because it was too big, but you can download it from here www.easygcc.com/correl.rar. On the sheet called " Correlation" there is matrix, I am baiscally trying to fill in the correlation formula into every cell so the matrix is filled out. The data for the correlation calculation comes from the sheet called "Tech Data". I have filled in the correlation formula into a few of the cells as an example, but I don't want to continue doing this manually but rather have a macro do it for me. Otherwise this will take for ever. My macro is also a part of the file, if you would liek to look at it and maybe fix what I already have.
View 4 Replies
View Related
Aug 21, 2006
The following code is supposed to produce six series on an xlXYScatter chart. It produces seven with the seventh series being a repeat of the sixth but named series 7.
Sub Chart2()
Dim DataRange As Range
Dim CellString As String 'Stores a cell range in the form "AA27:AB39"
Dim CurrentSeries As Integer
Dim SeasonCount As Integer
Worksheets("Hemisphere").ChartObjects(2).Activate
CurrentSeries = 1 ............
View 9 Replies
View Related
Jun 24, 2008
I have a list of data on one sheet and a 'reports' page on another sheet. The reports page has several pivot tables and a pivot chart. I want the pivot chart to format the bars on the chart relevant to the series name. The series names are "R" "A" amd "G" for Red Amber and Green respectively, I want the chart to change the colour of the series so that it is the correct colour ie. "R" would be coloured Red, "A" would be Amber and "G" would be green.
Sub PivotLoader()
Dim Red As Integer
Dim Amber As Integer
Dim Green As Integer
Red = Range("H9").Value
Amber = Range("H10").Value
Green = Range("H11").Value
Range("B8").Select
ActiveSheet.PivotTables("PivotTable4").PivotCache.Refresh
Range("D25").Select
ActiveWindow.SmallScroll Down:=18
Range("B49").Select...........................
View 2 Replies
View Related
Sep 27, 2011
I am trying to create a line graph that will incorporate multiple columns of data in one series of data. The reason I do not place all of the data in one column is because it could exceed the maximum amount of rows allowed in excel. Also I need the data split up for viewing purposes.
I can easily just graph one column but how do I combine all the columns into one line graph with the data being in separate columns. Basically all the columns will be my Y values and X values are just 1:n.
Example Below:
Column AColumn B Column C159261037114812
Now in the example all of the values are x values.
View 2 Replies
View Related
Sep 15, 2014
I have a table that I add rows to via a macro. I am trying to add correlation matrices that find the last row of the table, and include a certain number of rows above that to create a rolling correlation. For example, if I want the 2 week correlation, I would like the last 10 rows of data, but if I add another row, this formula should move down to incorporate the new row and deselect the now 11th day preceding. I do this manually but it is very time consuming.
View 9 Replies
View Related
Nov 18, 2009
I have a particular issue - i am trying to create a rolling correlation.
lets say you have two variables for 3 different months:
Month1 a x
Month2 b y
Month3 c z
correlation will correlate a-c with t-v
however, come next month - i have to manually drag the ranges for a rolling three month correlation:
Month1 a t
Month2 b u
Month3 c v
Month4 d w
correlation will now correlate b-d with u-w for a the last three months.
I want to automate this as I know what the month is, I can't seem to return the last three months and correlate the data.
I have tried to do this with the database functions, but I think that they break the array relationship of b-d and u-w in the correlation function.
View 9 Replies
View Related
Oct 17, 2005
I'm trying to Pearson Ranks correlation table. This is the formula I'm
using:
=PEARSON(Ranks!$B$3:$B$230,Ranks!B$3:B$230)
This is placed in cell B3. When I copy the formula over to DT3 the formula
correctly adjusts the B in the second part of the formula all the way to DT
and still references the values in the B column.
But I want to now copy this down to C4, D5, E6 and then over to DT for each
so that it correctly makes the correlation table.
View 12 Replies
View Related
Dec 17, 2006
I wonder if you think it can be possible to constraint the RAND function so that it gives off its result commensurate with the result of another RAND function with certain probability N? For example,
Say there are two random functions and the first generates 0,5667 the correlation is at 0,5 so there is a 50% probability that the second random function will generate the result very close to it, say, 0,6012….Do you think the second random function can be tweaked so that it performs in this manner….when there is a 100% correlation – the results of both random functions will agree…when it is 80% it will be very close…..do you think this can be done with NORMINV?
View 9 Replies
View Related
Nov 1, 2011
I have a combination chart with data from Sun thru Sat (7 days). My manager wants to display only the weekday data (M,T,W,Th,F) and remove the weekend data (Sat,Sun). However, I also need this data to be included in other charts I need. I was thinking in years past, that I was able to remove specific data from the chart directly without having to alter or change the spreadsheet data? I have copied a second set of data and hid the weekend data, however, I have gaps in the data, which I need to remove in the chart display. Not sure what to do, or the best avenue?
View 2 Replies
View Related
Aug 22, 2014
I consider myself decent at Excel, but I don't have quite the depth of knowledge I would like. Here is my issue:
I currently have a series of 50 numbers for each row on a 300 row spreadsheet. I would like to write a formula that identifies the six largest numbers in that series and adds them together. So, for example, if the series was 8-6-4-4-4-5-1-1-0-15-3-9, the formula would spit out 47 (15+9+8+6+5+4)
How do I write that kind of formula?
View 8 Replies
View Related
Sep 21, 2011
I would like to have two series of data using the same X axis (date, formatted in months). The Y axis is in intervals of 100,000.
The first data series is historical (actual) data (i.e. Jan 2009 to August 2011). The second data series is forecast (Sept 2011 - August 2012). So i want the forecast series to start immediately after the historical series. It is a 'line with markers' chart. The key objective is that the forecast data looks visually distinct from the historial series.
Excel version: Excel 2010
OS: Windows 7
View 5 Replies
View Related
May 15, 2008
I'm using the following code to set the colours of the slices of all the pie charts in my spreadsheet based on each slice's legend's label: ....
View 9 Replies
View Related
Dec 6, 2013
see attached pivot table... I believe that there is a statistical relationship between column B & C (together, i.e. their combined effect) upon column D. i.e. (B & C) - have some relationship to effect D.
(1) What formula would I use to measure this correlation?
View 3 Replies
View Related
Jul 29, 2013
I have a product list going down column A, Avg Sale Price List going down Column B, a Margin % List going down Column C, and a Correlation List going down Column D. Here is a small sample of my data.
Product Name
Avg Sale Price
Margin
Correlation
372 Grizzly Salmon Oil Total
48.99
35.31%
-0.423
[Code] ........
View 3 Replies
View Related
Jun 26, 2014
I am looking for the best way ( or choice of ways) to display the correlation between 2 lists of results.
I have a column of failure percentage rates for engineers that I want to compare against a second column of their compliance rate with a process. There are two levels also. The first set shows the good performers and their compliance, the second showing the poor performers.
My goal is to see whether their compliance with the process has any bearing on their failure rates. The full lists contain data on approximately 400 engineers in each category
(if the the numbers below are not in a proper format, I tried to download the excel converter thingy but the link failed)
faliure rate
Compliance
3.95%
91.40%
[Code]....
View 1 Replies
View Related
Jul 23, 2007
I have the monthly data stores column by column, e.g. Column B for Jan 07, Col C for Feb 07 etcetc, so every month I have a new column added to the end then I have some statistical analysis table which calculates such as annual return, standard deviation etcetc
How can I make a button which I can click then it can update to replace the formular with the latest column? for example, =CORREL(B2:B10, G2:G10), col B contains Feb 07 data and G contains June 07, let's say I add another column for July 07 which will be column H and I would like by clicking the button then all formula with col G will be replaced by col H?
View 8 Replies
View Related
Nov 26, 2007
I am facing a problem with WorksheetFunction.Correl
I got information, that matching class properties of Worksheetfunction is impossible.
This is my first experience with macro's and I hardly know how to deal with this problem.I got some excel file from my friend and tried to do this in the same way.
Macro was supposed to calculate the Pearson's coefficient of correlation for orders from 1 to 100 of autocorrelation in order to calculate Q Box-Pierce statistic.Data is prices of stocks for some bank and correlation coefficient is calculated for simple returns on stocks.
View 9 Replies
View Related
Apr 23, 2014
I am calculating multiple Correlations for long lists of data. I want the correlation based on the value in a given column. In the below example I would like a different correlation for each value in colA and the correlation will be between colC and colD. ColB lists the different people whose data are used for the correlations. In actuality there are 30 or more people for each test. In every case ColC is the same for a given value of ColB (although I doubt that matters). The number of people taking each test varies. I would like a correlation of ColC an ColD for each value of ColA.
Simplified version of my data
ColA ColB ColC ColD
Test1 Fred .75 1
Test1 Max .80 0
Test1 Sara .90 1
Test2 Fred .75 1
Test2 Max .80 1
Test2 Sara .90 0
....
Test100 Fred .75 1
Test100 Max .80 1
Test100 Sara .90 1
View 6 Replies
View Related
Jul 18, 2013
Any existing UDF that will take two ranges of Prices (for stocks) but get their respective returns to then get the correlation of that?
View 4 Replies
View Related
