I have some data which I would like to plot, presumably in a scatter chart. The data is in the following format.
Category, ID
Cat1, 560000
Cat1, 560005
Cat1, 880011
Cat2, 580000
Cat2, 580001
Cat2, 580002
Cat2, 780052
Cat3, 600000
Cat3, 600010
Cat3, 600011
Cat3, 1003452
For non-developers, you could think of this as Categories of Tyres and their Prices. I'd like to see that in the "Car" category, we sell items in a set of different price points and in the "Bicycle" category we sell items at a different set of price points. This metaphor only breaks down in that no 2 products can use the same price.
For developers out there, these are in fact a Long, Primary-Key to a particular relational database table and I'm attempting to plot how much and what parts of the available ID range have been utilised already, based on splitting the data into some categories.
I have 13 categories, so these will have to become the x axis due to excel limiting that to 255. Therefore ID will be my Y axis.
By using a 2D Line Chart on just the IDs, and squashing the chart, I was able to plot the overall usage of the ID range.
However, I'm unable to get any chart type to split this by category - I presume XY Scatter or Heatmap has this ability somehow.
Update:- Since posting I have encountered this "Contiguity Chart" which is roughly what I'm after but in Excel if feasible
https://qvdesign.wordpress.com/2012/03/29/qlikview-contiguity-chart-aka-that-chart-from-the-windows-disk-defragmenter/
Related
I am looking to plotting a car variant price comparison chart where the chart Y axis would be amount currency.
X Axis would be different car brands like Honda Hyundai, Toyota etc.
My plot would be the price for each car variant from low to high As per this image
Note:
Number of variants in the brands may be different so that way the series in the plots will be different in length/size.
I have a set of cost data for different pieces of unique equipment. Each piece of equipment is classified as a particular equipment class which I have pulled from an index match on the unique equipment number. I now have a set of ~9000 columns of cost data, each with a column header of one of the ~300 equipment classes.
What I want to do is to get the median, 25%, and 75% for the full data set for each of these equipment classes.
I either want to create a single long column of all the data for each equipment class, or have a way to calculate the Percentile() values for the data in all columns with the same heading.
I could filter the data for each equipment class one at a time and calculate the percentile values, but with 300 equipment classes it would take forever.
Example:
Class01 Class02 Class01 Class03 Class03
1 4 7 10 13
2 5 8 11 14
3 6 9 12 15
And I want the 25%, median and 75% for the distribution for Class01, Class02, and Class03
Thank you for your time.
I either want to create a single long column of all the data for each equipment class, or have a way to calculate the Percentile() values for the data in all columns with the same heading.
I'll just tell you how to change your data around. From there the percentiles/ quartiles will be straight forward.
Start with your data like this. Notice that I added a column on the left. It's easy to make, just type Item1 and drag down (or double click the small square in the bottom right corner of the cell)
You then need to hit Alt+D+P.
Select multiple consolidated ranges > next
Next (create page ranges for me...)
Select all of your data as the range, click add then finish
You will now get a pivot table that looks like this:
Click the grand-grand total (i.e. 120) and that will create another pivot table like this:
Et voila...
I am working with Excel and Powerpivot. In my data layer I have two tables related, in the main table I have my data and the ID that refers to my second table which is a reference table containing (Description, short description, units, abbreviation,..)
Simplified example:
Table 1:
Name Year Value Indicator
Usain Bolt 2009 9.58 1
Mike Powell 1990 8.95 2
Table 2:
ID Discipline Short_Descr Unit
1 100 metres 100m seconds
2 Long Jump L. j. metres
I want to create a chart where the axis and title are dynamic using the values in the reference table. The dynamic title is quite easy to build: I create my pivot chart using as Report filter the Description, and then I reference the chat title to this filter. But I would like to include in the vartical axis dinamically the units. So if I change the filter the units in the axis also change
I am doing a project that requires me to study the several condition that affect GPS accuracy, and after I collected a set of data and dumped it to Excel, I was trying to plot a scatter graph, grouping the data into different series according to a value: in this case, I wanted to plot the Latitude and Longitude values as the XY scatter values, and separate the series by the number of satellites when the fix was obtained.
Timestamp Latitude Longitude #Satellites
133009.279 3839.3354 904.7395 0
133010.279 3839.3354 904.7395 0
133011.279 3839.3354 904.7395 0
133026 3845.9863 907.4513 4
133027 3845.986 907.4491 4
133028 3845.9851 907.448 4
133222 3845.9909 907.4866 4
133023.28 3845.9817 907.4429 5
133024.28 3845.9867 907.4549 5
133048 3845.9868 907.452 5
133205 3845.9929 907.4858 5
133206 3845.9927 907.486 5
133207 3845.9925 907.4862 5
133056 3845.9885 907.4569 6
133057 3845.9881 907.4578 6
133223 3845.9905 907.4868 6
133224 3845.9901 907.487 6
I have tried selecting the three rows, adding the series afterwards by selecting the appropriate row, and even tried pivot tables, but these don't allow for scatter-plots unfortunately.
All this to no avail, but I am positive that you can plot the graph. Does anyone have an idea?
PS: Manually selecting the series myself isn't an option, since there is a large number of data. If I could select all of the data for one specific value in a row, though, would let my select each series, and I think I would be able to make it from there.
Have a look at the XY charts from FusionCharts XT - http://www.fusioncharts.com/demos/gallery/#bubble-and-xycharts
I seldom if ever use excel and have no deep understanding of graphs and graphing-related functions. Having said that...
I have dozens of rows of data, composed by 4 columns
column 1 = amount/price (in numbers)
column 2 = description (the what
in text)
column 3 = category (in text)
column 4 = subcategory (in
text)
I want to make a bar graph of my rows of data so that, the end result looks like this:
X axis - categories
Y axis - amount/price
The trick here is for categories NOT to repeat themselves. For example, if our data is something like...
100 | boat purchase | boats | 3 engine boat
200 | boat purchase |
boats | 2 engine boat
500 | plane purchase | planes | 4 engine plane
900 | car purchase | cars | 1 engine car
Then there should only be ONE instance of boats, planes and cars in my graph, under which all associated data would be summed up.
Last but not least, I have seen graphs where, these unique-not-repeated categories, instead of just being one single 'bar' so to speak, are composed of smaller bars. In this case, I want these smaller bars to be the sub categories, so that the end result would look like this:
In that sample image, I first present a 'basic, classic' graph where blue, yellow and red each represent a unique, different category. Right below it is what I want, a 'breakdown' of each category by subcategory where blue/yellow/red each represent an imaginary 3 different subcategories per category.
This means subcategories will repeat themselves for each category, but categories themselves will not.
For clarification, I currently only have 3 main categories and 6 or so sub-categories, but this could change in the future, hence the desire to have this in an automatic/dynamic fashion
Kind regards
G.Campos
EDIT: new image:
Here i my take on it. Unfortunately I can't post the screenshots as I don't have enough posts.
One solution is to use pivot charts put Amount in "Values", Category in "Row Lables", and SubCategory in "Column Labels".
I uploaded relevant images on a free image upload service.
This is our source data:
Amount Decription Category SubCategory
100 boat purchase boats 3 engine boat
200 boat purchase boats 2 engine boat
500 plane purchase planes 4 engine plane
900 car purchase cars 1 engine car
450 boat purchase boats 2 engine boat
110 plane purchase planes 4 engine plane
550 car purchase cars 1 engine car
230 car purchase cars 2 engine car
450 car purchase cars 5 engine car
This is the desired graph (Edit: This has ghost bars):
http://imageshack.us/photo/my-images/849/pivot.gif/
I just read the comment about no ghost graphs. This might be what you are looking for:
http://imageshack.us/photo/my-images/266/pivotnoghost.gif/
Just googled and found something very similar here:
peltiertech.com/WordPress/using-pivot-table-data-for-a-chart-with-a-dual-category-axis/
You need to add http:// ( I can't have more than two hyperlinks due to low number of posts)
I am not sure this will get you exactly where you want but I find in general in excel it is easiest to summarize your graph data on a separate tab.
For sample data like this
you would create a 2nd tab in the sheet that appears something like
the totals are calculated by using the sumif formula
=SUMIF(Data!C:C,Summary!A2,Data!A:A)
For the Category totals
and
=SUMIF(Data!D:D,Summary!E2,Data!A:A)
For the sub category totals (Assuming sub-categories are mutally exclusive). Now that that data is summarized, highlight the cells and insert a column chart for the following charts.
Adding new categories and/or sub categories will require you to add lines to the summary data, and then add series to the charts. You could use a vba macro to automate that task but I suspect that is overkill since your dataset is "dozens" rather than "thousands"