Calculating linear regression with DAX in a Tabular model

Hi,
I'm interesting to know if anyone has faced the linear regression in a Tabular model by DAX formulas.
Thanks

It's been a while, but... recently Rob Collie did a blog post about this
here.

Similar Messages

Graphing Linear Regression With JFreeChart

Hi everybody I'm making a school project where I need to graph linear regression and scatter diagram in the same graphic, I am using JFreeChart and is a little confusing, I have already made a basic graph but I want to know if someone can tell me how to draw the line for the regression, i have de the values of a, b [bx + a], and also how to make the thickness of the line more thin...Here is my code any ideas are welcome :D
        JFrame frm_grafica = new JFrame("Grafica - B-Rabbit");
        frm_grafica.setVisible(true);
        frm_grafica.setSize(250, 250);
        frm_grafica.setLocation(this.getLocation().x+this.getWidth()+5, this.getLocation().y);
        XYSeries serie1 = new XYSeries("Diagrama de Dispersion");
        XYSeries serie2 = new XYSeries("Regresion Lineal");
        for(int i = 0; i < limitRows; i++){
            serie1.add(matriz[0], matriz[i][1]);
for(double i = 0; i <= limitRows; i+=0.1){
serie2.add(i, b*i+a);
XYSeriesCollection datasetCollection = new XYSeriesCollection();
datasetCollection.addSeries(serie1);
XYSeriesCollection datasetCollection2 = new XYSeriesCollection();
datasetCollection.addSeries(serie2);
XYDataset dataset = datasetCollection;
XYDataset dataset2 = datasetCollection2;
JFreeChart chart = ChartFactory.createScatterPlot(
"Diagrama de Dispersion",
"X",
"Y",
dataset,
PlotOrientation.VERTICAL,
false,
true,
false
XYPlot plot = (XYPlot)chart.getXYPlot();
plot.setDataset(1, dataset2);
XYLineAndShapeRenderer renderer = (XYLineAndShapeRenderer) plot.getRenderer();
renderer.setSeriesShapesVisible(0, true);
renderer.setSeriesShapesFilled(0, true);
NumberAxis rangeAxis = (NumberAxis)plot.getRangeAxis();
rangeAxis.setStandardTickUnits(NumberAxis.createStandardTickUnits());
ChartPanel chartPanel = new ChartPanel(chart);
frm_grafica.add(chartPanel);
chartPanel.setVisible(true);

I think I have an idea as to what to do. I still need to retain the integrity of the original data, therefore I will be creating new channels.
In a script you could do a loop looking for a value that is within the range then copy that data set to the new channel. Are there any builtin commands that would make this process quicker or more streamlined? All I know to do is to look at each value with CHD and make the decision from there.
Thanks,
Wayne

Updation of DAX in SSAS Tabular Model

Hi All,
I have a SSAS tabular model which has around 42 Dimension table and 10 Fact table each containing some thousands
of record.when I am writing some DAX on a FACT to get some measure ,its taking 10-15 minutes time to update one DAX.
I have to write more than 300 measures using DAX.Can anyone please suggest how can I speed up the updation process
in writing DAX.
Thanks in Advance.
Sanjay
sanjay

Hi Sanjay,
According to your description, there are around 42 Dimension table and 10 Fact table each containing some thousands of record in your tabular model, so it takes 10-15 minutes time to update the calculation to server, right?
It's default setting to update the modification to server when change something on your tabular model, if the tables contain large data, it will takes long time to update. In your scenario, we recommend you to use Multidimensional database instead of Tabular
model when having large amount of data with complex requirements.
SQL Server Analysis Services (SSAS) Multidimensional model is used when having large amount of data with complex requirements. In order to improve query performance, there will be a cube doing heavy time-consuming processing and then synchronize it
to query cubes.
However, SSAS Tabular model is used when the data model is relatively simple. So for the large amount data model, we recommend to use Multidimensional database instead of process data on one server then synchronize it to query cubes.
Reference
http://blog.aditi.com/data/choosing-between-analysis-services-multidimensional-and-tabular-models-part-3/
Regards,
Charlie Liao
TechNet Community Support

Cell Coloring using DAX in SSAS Tabular Model

HI All,
I need to implement background and foreground cell coloring in SSAS Tabular Model using DAX or any other feasible solution similar to what Cell Properties does with MDX which is supported by SSAS Cubes.
Request the gurus to help me with this problem.
Thanks
ET

Hi Et,
As Marco said, this feature is not supported in current version. If you have any concern about this behavior, you can submit a feedback at
http://connect.microsoft.com/SQLServer/Feedback and hope it is resolved in the next release of service pack or product. Your feedback enables Microsoft to make software and services the best that
they can be, Microsoft might consider to add this feature in the following release after official confirmation.
Thank you for your understanding.
Regards,
Charlie Liao
TechNet Community Support

Linear regression calculation

Need to know whether an algorithm to calculate linear regressions exist, given a determined number of points.

Dear Ingo,
I am not sure about your answer. Well, I need an script as a solution. Imagine, the time data is collected in a channel, DatCha_1. There are 1000 points corresponding to 10 seconds. I have the corresponding aceleration data, in Dat_Cha_2, that is, 1000 acceleration points. So, I need to calculate the linear regression with this acceleration data each second, that is, I must obtain the a, b data of a y=a+bx corresponding to each 100 values of the acceleration data, so 10 a values and 10 b values. I must divide the original acceleration data into 10 interval and obtain the linear regresion of each of this intervals.
What should be the script for this action? You mention to adjust the channel length of the time and data channel, but I do not know what you mean. In fact the function is:
Call CHNREGRYCALC("Time","Acceleration","Time","Regress_Y","linear") '... X,Y,XCHNNO,E,REGRTYPE
Could I use a new time channel only with 10 values and the original aceleration with 1000 values?
So, please could you send me a little example of the script I would need to solve the problem?
Regards,
Koniker

Linear regression

Hi,
I would like to know how I could do the following linear regression calculation. I have two data channels, the first one corresponding to time information and the second one about aceleration. I must do linear regression with this acceleration data each second, but taking into account that the time channel has 10585 points for example for 44.1 seconds, corresponding to a 240msec sampling frecuency. I suppose that I must use the ChnRegrXYCalc function but I would like to know what the parameters should be.
REgards
Koniker.

Dear Ingo,
I am not sure about your answer. Well, I need an script as a solution. Imagine, the time data is collected in a channel, DatCha_1. There are 1000 points corresponding to 10 seconds. I have the corresponding aceleration data, in Dat_Cha_2, that is, 1000 acceleration points. So, I need to calculate the linear regression with this acceleration data each second, that is, I must obtain the a, b data of a y=a+bx corresponding to each 100 values of the acceleration data, so 10 a values and 10 b values. I must divide the original acceleration data into 10 interval and obtain the linear regresion of each of this intervals.
What should be the script for this action? You mention to adjust the channel length of the time and data channel, but I do not know what you mean. In fact the function is:
Call CHNREGRYCALC("Time","Acceleration","Time","Regress_Y","linear") '... X,Y,XCHNNO,E,REGRTYPE
Could I use a new time channel only with 10 values and the original aceleration with 1000 values?
So, please could you send me a little example of the script I would need to solve the problem?
Regards,
Koniker

Creating Tabular Models using AMO

Hi,
I am trying to setup Tabular Models on SSAS using AMO referring to the materials available at Technet and the Codeplex AMO Tabular Tutorial. I have a few questions/issues
currently and was hoping I could get some help from here. The tool I am working with currently supports Multidimensional mode via AMO and I am trying to extend most of its features from the Multidimensional model. So, some of the questions might not exactly
fit into the Tabular model and if so, please do point them out.
RELATIONSHIPS: While creating relationships among tables,
would it suffice to create the relationship in the Data Source View Schema (as the tool currently does for the Multidimensional mode) or should we explicitly create relationships under the dimensions that we create to represent the table in Tabular mode? The
latter is the approach mentioned in the tutorial and I wanted to know if there is any difference between the relationships created using the 2 methods.
MEASURES: I am trying to create measures using Native
OLAP Measure objects in the Tabular model. I understand the shortcomings of this type of measures but I would still need to create them for basic aggregation functions like SUMMATION, COUNTS etc… I tried creating the measures using the same Measure Group that
was created to represent the table and used a ColumnBinding to the Measure’s source-column for the measure object’s source property but I get the following error:
Binding for VertiPaq measure MEASURE_NAME in measure group MEASURE_GROUP_NAME is invalid because it does not match any property binding of the fact dimension
TABLE_NAME
Am I missing anything here? Is there a better way of creating OLAP measures in TABULAR model without using the default Table dimensions that we create?
HIERARCHIES: I tried creating new Dimensions to hold
hierarchies for a table but when creating them, I get the error about MISSING ROWNUMBER attribute. Is this attribute mandatory for every dimension that is created?
To avoid this problem, I used the same dimension that was created to represent the table and tried adding inter-table attribute relationships to it but I get the following exception
message which I cannot figure out.
VertiPaq property ‘’ cannot have a name binding.
In general, can we create separate measure groups and dimensions, apart from the ones we use to represent the table, to store the custom measures and hierarchies? Is this a recommended
approach? This way, I am trying to keep things in parallel with the Multidimensional model that our tool currently supports but when I create such individual dimensions and measure groups, I get an error on the mismatch between the number of measure groups
and dimensions in the table.
Please bear with the long list of questions. I could not find any help online for these and so am posting them all here.
Thank you.

RELATIONSHIPS: While creating relationships among tables,
would it suffice to create the relationship in the Data Source View Schema (as the tool currently does for the Multidimensional mode) or should we explicitly create relationships under the dimensions that we create to represent the table in Tabular mode? The
latter is the approach mentioned in the tutorial and I wanted to know if there is any difference between the relationships created using the 2 methods.
No - relationships in the DSV have no impact on the end model. You need to explicitly create relationships between your dimensions and measure groups for them to be picked up as relationships in your tabular model.
MEASURES: I am trying to create measures using Native OLAP
Measure objects in the Tabular model. I understand the shortcomings of this type of measures but I would still need to create them for basic aggregation functions like SUMMATION, COUNTS etc… I tried creating the measures using the same Measure Group that was
created to represent the table and used a ColumnBinding to the Measure’s source-column for the measure object’s source property but I get the following error:
Binding for VertiPaq measure MEASURE_NAME in measure group MEASURE_GROUP_NAME is invalid because it does not match any property binding of the fact dimension TABLE_NAME
Am I missing anything here? Is there a better way of creating OLAP measures in TABULAR model without using the default Table dimensions that we create?
There is no concept of native measures in a Tabular model. You need to create all your measures as "Calculated Measures" in AMO, but using the appropriate DAX expressions instead of MDX.
HIERARCHIES: I tried creating new Dimensions to hold hierarchies
for a table but when creating them, I get the error about MISSING ROWNUMBER attribute. Is this attribute mandatory for every dimension that is created?
To avoid this problem, I used the same dimension that was created to represent the table and tried adding inter-table attribute relationships to it but I get the following exception message
which I cannot figure out.
VertiPaq property ‘’ cannot have a name binding.
Yes, I believe every table needs the hidden RowNumber attribute. The
TableAddEmptyTable function in the tabular AMO sample on codeplex shows you how to create this.
In general, can we create separate measure groups and dimensions, apart from the ones we use to represent the table, to store the custom measures and hierarchies? Is this a recommended
approach? This way, I am trying to keep things in parallel with the Multidimensional model that our tool currently supports but when I create such individual dimensions and measure groups, I get an error on the mismatch between the number of measure groups
and dimensions in the table.
No you can't create any extra structures. Tabular projects only supports a subset of AMO. You need to follow the example on codeplex very closely and read all the code comments if you are making changes because it's very easy to break things.
My suggestion is to create an abstraction layer either using the TabularAMO library from codeplex as it is or creating your own library if you only need a subset of the functionality. This will mean that your core code is not too tightly bound to AMO. The
reason for this is that I would hope that MS will replace AMO with something better for Tabular models in a coming release and having a clear abstraction layer should make it easier to update to a new API.
http://darren.gosbell.com - please mark correct answers

How can I make a graphic (scatter chart) with 1 "y", but multiple "x" (to make a multiple linear regression)?

how can I make a graphic (scatter chart) with 1 "y", but multiple "x" (to make a multiple linear regression)?

Émilie,
The default for X-Y Charts is for X values not to be shared, so it's entirely possible for you to create X-Y pair sets that have common Y-values and independent X-values.
This answer assumes, as always, that I properly interpreted your problem statement. A linear fit line for each pair set should satisfy.
Jerry

Time Intelligence DAX functions are not working in Tabular model

Hello,
I've implemented a simple tabular model using SQL Server 2012 SP1 SSDT with a date dimension that has an active relationship with the date key in the fact table and has met the criteria for a date dimension to the best of my knowledge, i.e. no missing
dates, etc.
However, when I add a calculated measure in the VS project using any time intelligence function, i.e. =CALCULATE(SUM([NewPatientTotal]),TOTALYTD(DimDate[Date])), the calculated measure only returns the value of the current member, i.e. SUM(NewPatientTotal])
without throwing an error. This is happens for all time intelligence functions!
Are there any best practices that I've missed? I've scoured the SSAS forums to no avail.
Thanks,
Ben
Ben Lezin

You probably forgot to use the "Mark as Date Table" on the DimDate table.
Marco Russo (Blog,
Twitter,
LinkedIn) - sqlbi.com:
Articles, Videos,
Tools, Consultancy,
Training
Format with DAX Formatter and design with
DAX Patterns. Learn
Power Pivot and SSAS Tabular.

How to create Date relationship with Fact and DimDate on Tabular model?

I have two tables in SQL Server 2014 SSAS Tabular model.
DimDate(defined as data table) have date column (format is 1.7.2010 00:00:00) This table have been impoted from AdventureWorks.
FactSales have SalesDate column (format is 25.03.2015 18:08:05). This table have been imported from Excel. Colums is defined as date in Excel.
When I set Data Type as date in Tabular I get error:
"Datatype conversion failed for table...Value:'25.03.2015 18:08:05'"
I have tried to create calculated colums like =DATEVALUE([SalesDate]) and =DATE(YEAR([SalesDate]),MONTH([SalesDate]), DAY([SalesDate])), but getting error.
What should I do so that I can create relationship between Date column?
Kenny_I

Hi Kenny_l,
According to your description, you fail to convert the data type into date after creating calculated column with expression. Right?
In Analysis Services, the tabular will detect the source data type to determine which data type can be converted to. If you want to convert a data into date type, no matter this data is from database or a file, you must make sure this data can be recognized
as a date in SQL Server. In this scenario, the format '25.03.2015 18:08:05' can't be recognized as date even you apply DAX functions. So please change the format of the column in excel like "03.25.2015 18:08:05".
PS: For testing, you can create a temp table and insert sample text into a date column, because the date format in SQL Server depends on the location selection when installing.
After retrieving data from data source and all values within column are correct format, you can change the data type into Date, and select the format you expect.
If you have any question, please feel free to ask.
Best Regards,
Simon Hou
TechNet Community Support

SSRS with DAX - parameters multivalue vs total element

Hi all,
I'm using SSRS on top of a SSAS tabular model to create intuitive dashboards for management department.
One thing I'm struggling with is the lack of a "total" on parameters. What I'm doing is, based on the fact table, I create the datasets that will feed each parameter, so it only has the dimension members available on the fact table. That is fine
performance wise, you only have 1 element and it renders very quickly.
For a management perspective at a higher level it is important to view the totals to have the big picture of the information. In my project we are in a University, so I have the parameters Year, school (there are several schools in the university campus)
and the course (computer sciences, math, statistics, etc).
At first glance, if I'm a course coordinador I want to pick my course and have a global view, and I'm achiveing that, but imagine now, that I am the school coordinator, or the university coordenator.
I want to see the aggregate values by all courses or by all schools.
In my view I have 2 ways to do this and I come from a tradition cubes background:
1 - I would use the all element the cubes hierarchy have on each attribute
2 - I would use multivalue parameters with StrToSet (worse performance)
Using DAX what is your approach?
Best regards

Hi AsifMehmood,
Per my understanding you have create an SSRS report with SharePoint list, now you don’t know to create the distinct parameters by using CAML query, right?
For the CAML language doesn’t have any reserved word (or tag) to set this particular filter to remove duplicate results, but we can use the custom code to do this function. I have tested on my local environment and we can do that by create one hidden parameter(Param1)
to get all the values from the fields which will add the filter and then create another parameter(Param2) to get the distinct values based on the Param1, we use the custom code to do the deduplication.
Step by Steps information in below thread for your reference to create the parameters and the custom code:
"How to get distinct values of sharepoint column using SSRS"
Other similar thread for your reference:
https://audministrator.wordpress.com/2014/02/17/sharepoint-list-add-distinct-parameter-value/
If your problem still exists, please feel free to ask and also try to provide us more details information.
Regards
Vicky Liu

How to make a report to display next 18 months of data with when user select a particular month from the filter in power pivot tabular model.

Hi,
i have a dimension table with month_key having values (201201,201202,201203.......202011,202012) and month name ( Jan 12, feb 12,......NOV 20, Dec 20) and a fact table with columns (month_key ,measure_types, Amount)
My requirement is to create a power pivot report in which when a user select a month from the filter, the report should display the (selected month+18 ) month's data against each type . when JAN 12 is selected ,the jan 2012 +18 = june 2013
, month name should be populated with months till june 2013 only .
i tried creating calculated column"END DATE " in the fact table with dax expression to calculate the 18th monh from the current month as below
month_key END DATE
201201       201306
201202       201307
and thought of filtering the table with month key <= ENDDATE but it is not working as expected. could you please guide me on this ? Is there any time intelligence function that serve the purpose . Iam using excel 2010
..hence could not do any calculation on the report side also. please suggest .
Thanks in advance

Do you need to show the measure calculated for those 18 months as a total on 1 row, or do you need to select a single month and then display on row filters 18 distinct rows?
The first is trivial as driezl has suggested.
The second will require a second calendar table.
I created this example workbook for a coworker who had a similar problem. You will have to use the disconnected table as your filter and pull your related table onto the rows.
Finally, the easiest way to deal with the sort of date arithmetic you need to do is to restructure your date table to have a series of "Sequential" fields. These fields should be the number of units of time since the beginning of your calendar.
For example, consider a calendar starting on January 1, 2010. For January - December 2010, [MonthSequential] = 1, 2, ..., 12. For January - December 2011, [MonthSequential] = 13, 14, ..., 24, and so on, incrementing by 1 for each sequential month in time.
Assuming you have this set up in your date tables (one related to your model - DimDate - and one disconnected - DisconDimDate) your measure would look like this:
18 Month Measure:=
CALCULATE( [Measure]
, FILTER( DimDate
, DimDate[MonthSequential] >= MAX( DisconDimDate[MonthSequential] )
&& DimDate[MonthSequential] <= MAX( DisconDimDate[MonthSequential] ) + 18
Please review this example along with the workbook I have linked above.

SSRS report with tabular model – MDX query CoalesceEmpty function does not return the provided string value

Hello everyone,
I created following calculated member in MDX query. I am using it in one of the report parameter in dataset (single select dropdown list as report parameter).
WITH MEMBER [Measures].[ParameterCaption] AS
CoalesceEmpty([Customer].[National Account Code].CURRENTMEMBER.MEMBER_CAPTION,'None')
I would like to display 'None' text at the top of the dropdown list values. So that when user selects 'None' then this parameter will not considered in MDX query else the selected National Account Code will be considered to filter report data. But,
the above return blank/empty value for [Customer].[National Account Code].& member though I specified 'None' as text in CoalesceEmpty function. Any advice would be appreciated.
Thanks, Ankit Shah
Inkey Solutions, India.
Microsoft Certified Business Management Solutions Professionals
http://www.inkeysolutions.com/MicrosoftDynamicsCRM.html

Hi Ankit,
It seems that you issue had been solved in your another thread.
http://social.msdn.microsoft.com/Forums/sqlserver/en-US/5a5becac-226f-428a-95b0-aaaa22733818/ssrs-report-with-tabular-model-create-a-dropdown-report-parameter-with-none-option-as-the-top?forum=sqlanalysisservices#0e51bf8c-a66c-4df5-a244-0147728fdfdb
iif([Customer].[National Account Code].CURRENTMEMBER.MEMBER_CAPTION="","None",[Customer].[National
Account Code].CURRENTMEMBER.MEMBER_CAPTION)
I marked this reply as answer, it will benefit to other members who have the similar issue.
Regards,
Charlie Liao
TechNet Community Support

Error about using a concatenated calculated column in a SSAS Tabular model

Hi,
in a my tabular model I've added in a dimension table a calculated column concatenating a code and a description. The code is used to create a relation between a fact table and this dimension table.
When I try to use the tabular model inside Excel and select the concatenated column as a filter I've this error message:
I've tried to recalculate some times the tabular database after the deployment of the model changes and I've tried to run a full process of the entire database but any results.
Any suggests to me in order to solve this issue, please?
Thanks

Hi, I've solved. The concatenate formula was with "+" operator and not "&".
But during the column creation I've any errors also if the model in SSDT was empty.
Bye

Seasonal linear regression

Hi there, APO Guru's
We have created a forecast profile that uses strategy 35.
However when we run the forecast with this model for say the last 36months of history. The system proposes an almost straight line forecast. If we run it with 18 months history then the system proposes an acceptable forecast. I can now clearly see the sales history pattern in my forecast.
We do have enough data and seasonality in our history for the 36 months to work.
Can anyone maybe shed some light on this.
We also recently upgraded to SCM 5.0
Thanks for your assistance
Robert

Before the system applies seasonal linear regression, it carries out a seasonal test. The system uses this test to check if the historical data contains any seasonal patterns. For this, the system determines the autocorrelation coefficient for all periods.
If the value determined is at least 0.3, the system applies seasonal linear regression. If the value is less than 0.3, the system does not recognise a seasonal pattern and applies linear regression.
So I think the test against your 36 months of data returns a value of less than 0.3 and hence it uses linear only, and the value for your 18 months seems to be greater than 0.3.
(P.S. You can change this limit in the /SAPAPO/SCM_FCSTPARA BAdI Method PARAMETER_SET)

Calculating linear regression with DAX in a Tabular model

Similar Messages

Maybe you are looking for