Hi I'm trying to use the built in forecasting function in tableau and I think it's greyed out because my timeseries values have gaps. Basically there are hours during the day when nothing happened in my data, so nothing got recorded. I'm trying to build a forecast but my data jumps to the 2nd hour, then has results for each hour up to the 6, and jumps to the 9th. I know in ssrs I would try left joining the sql query I'm using as a datasource to a derived field with all 24 hours. I was hoping there was a slicker trick in tableau. All help is greatly appreciated.
Without having your worksheet, I'm not sure if it could be helpful, but have you already tried this quick fix suggested by Tableau?
Basically you should use discrete measures and using the "Fill missing values with zeroes" option.
Related
I am trying to move the old excel to qlik sense, there is a graphic that is made from the previous day. Then that graph has the dimension of date and measurement has the sum of the amounts per month.
So filtering this by day can be done with a filtering panel, but clicking every day to show me everything can be annoying.
My idea was something like a box where you can enter a value, for example a 6, so when typing a 6 I should show the graph from day 1 to 5.
That is my problem, how could it be done? where can I enter some expression or something to the dimension to calculate the days I want in a simple and intuitive way.
Another idea I had, was to create in qlik sense when preparing the data, a new column that content the sum of the amounts of each day, so that value that I want to show would accumulate
What do you recommend? any help is welcome thanks
I am trying to build a forecast for interest expense for floating debt in my company.
I have been given a set of ResetDates which help me match a given rate based on when the ResetDate is.
I have been successful in forecasting one period, but I need a much longer set of periods to satisfy my requirements.
I've tried derive nodes and nested if statements as well as filler nodes.
I am given this data to work with, I can only look at one ResetDate ahead.
Here you will find the data I used: Columns A/B/C/D is what i'm given, Column E (or 5th column from left to right) is what I want to derive as my output
I want to use 'InterestPayDate' and derive:
if it's more than 'NextReset' , the add 90 days to the 'NextReset' to create 'NextReset2'
That is as far as I can get.... where my problem lies is I want to look at NextReset2 and derive:
if 'InterestPayDate' is more than 'NextReset2', then add 90 days to 'NextReset2', if it's less than 'NextReset2', keep the current value for 'NextReset2'
Output should look like Column E here
Not sure if I need to dig deeper into the logical functions, in all honesty, I've just picked up SPSS and I am really trying to learn. Hopefully, you can point me in the right direction.
Thank you.
After computing the first NextReset2, you need to use a Filler node like the one below to change the value of the field.
You might need more than one identical nodes like this - one for each potential 90-day period that you are looking to extend the NextReset2 date. In your sample data, you will need at least two Filler nodes to get the correct value of NextReset2 for the last of the records.
There might be a more elegant way to do it, but this will work and it's easy enough to make copies of a node and string them together like this.
Please also see a sample IBM SPSS Modeler stream showing this approach here and using your sample data.
I am trying to calculate the missing weeks in the data. Below image shows how is my existing data.
The below image shows what is expected results:I am trying to calculate "Calculated Calendar Week" and "Calculated week Diff"
Any help in achieving the desired results is greatly appreciated!!
Thanks,
Ganesh
I fixed this by creating another data set with all the missing weeks and did a right join!
I'm calculating how long it takes to detect an issue in Tableau.
I have the following calculated field to work out mean time to detection:
DATEDIFF('hour',[DATE Reported], [DATE Responded])
When I use the meantimetodetect calculated field it only shows me one result and the rest 0s.
The one result shown is correct there were 3 days apart from when it was reported to the response date. The rest of the results are reported and responded in the same day so I dont know if that has anything to do with it?
Does anyone know why it is displaying like this perhaps there is a better way to calculate it?
Thanks.
I know we can generate chart from excel/csv data using ms chart library. But my question is little bit different. I am just asking for possibilities.
Is there any way to get the data from excel chart?
For those who wants to know what i did. I search it on google but no related result found. I don't know where to start. So, no code written at the moment. I know how to read excel file in powershell.
Any suggestion.
Thanks for your time.
What you are looking for is data or rather data elements.
What is a chart? If we think about a chart: It has an X and Y axis:
It has for the sake of argument and X-axis and we can call that date or data points.
It has a Y-axis for say quantity. Your manager may ask, show me how many logins are done per day for the last 30 days?
You would map out by day or by week how many logins were done and to show this in a "pretty" way: you can do this with a chart.
So, back to your question:
You need to know - What scale are you dealing with? Days, weeks, months, number of readings, etc.
You need to know - What quantity was taken at that time: Days. weeks, months, number of readings, etc.
Once you have both of these, you can then build a chart. Otherwise, using your Google example: It is just a picture and would be meaningless if there is not thing to tie it for a data relationship.
Hope this helps?