I have built an analysis cube and I am wondering how I can now somehow add in some kind of forecastinbg to it?
I have a time dimension in years, quarts, months, days.
Is there a way I can show forecasting calculations for future dates?
Hi I installed the add-in for excel data mining but when i try to select a table, in the tab of analyze, dont have any properties. please help me.
carlos of southamerica
I am doing a search on a column of type varchar, the columns all contain dates in the future. When i perform a query
SELECT [CIFPan]
,[CIFMemNum]
,[CIFLName]
,[CIFFName]
,[CIFExpDate]
FROM [FutureSoft].[dbo].[zCIFRecord]
WHERE CAST([CIFExpDate] AS smalldatetime) between '12/01/2000' and '01/30/2015'
I get error:
Msg 295, Level 16, State 3, Line 1
Conversion failed when converting character string to smalldatetime data type.
I can use this same query on an exactly same other field that contains dates in the past. Is there some kind of limitation on the dates as far as them being in the future or past?
how can use this mdx script in the calculation part of a cube, will i simply dump it in the script form by starting with the 'create member current cube.
[measures].[test]'
selectÂ
[measures].[abc] on 0,
[xyz].[xyz].(&0):[xyz].[xyz].(&60) on 1
from
(
select
(tail([month].[month].[month].members,6))on 0
from
[cube])
I am looking a way to use a tail funcion to achieve something like this
Â
If you see the picture i want obtain in one row the total amount of the month over a selected range. If I select months between 1 and 12, i want the values only for the month 12, if a select months between 1 and 10 I want only de value for October and so on.
Â
I tried this simple mdx query:Â
Select
Tail([Id Fecha Orden Pago].[Mes].allmembers,1) on
Rows
,
{[Measures].[Total Gastos Orden Pago]} on
Columns
From
[CUB_DGM_TestEnamiDW] WHERE
[DIM Programa Fomento].[Nombre Programa Fomento].&[RECONOCIMIENTO DE RESERVAS]
I am currently creating calculations for weekly and fortnightly ranges by day.
Is it better to use LastPeriods or Tail? For example, the last 14 days could be:
LastPeriods(14,Â
StrToMember
(
"[Date].[Year Week Day].[Date].&[" + FORMAT(Now(),"yyyy-MM-ddT00:00:00")
+"]"
) Â Â
))
Tail(
Filter(
{[Date].[Year Week Day].[Day]},Â
([Date].[Year Week Day].CURRENTMEMBER, [Measures].[Test Measure]) > 0
), 14
)
What is the advantages or disadvantages?
I am working with a data set containing several years' of monetary values. I have entries for past dates and the associated values, and I also have entries for future dates. I need to populate the values of the future date records with the values from the same date the previous year. Is there any way this can be done in Power Pivot?
View 6 Replies View Relatedhello
how can i select all dates upto todays date and include the first next future date in a given data base
say todays date was the 01/06/2006 (MM,DD,YYYY)
below is a mock data base
id date (MM,DD,YYYY)
1 01/02/2006
2 01/04/2006
3 01/06/2006
4 01/09/2006
5 01/20/2006
i want to select all dates equal or less that 01/06/2006 and include the first next future date .. and in this case it would be 01/09/2006
so the results would return
1 01/02/2006
2 01/04/2006
3 01/06/2006
4 01/09/2006
I am trying to compute average of Sales amount for 10 days and I am having an issue when using LAG function. Since weekends fall in between working days my LAG (10)Â function is also including weekends and hence I am not able to get the correct average.
For ex: D1 - 10 , D2 - 20, D3 - 30, D4 - 50, D5 - 10, D6 - W, D7 - W, D8-10, D9 - 20, D10 -30, D11 - 10, D12 - 40, D13 - W, D14 - W
I am using a sample script
SUM([Time].[Hierarchy].CurrentMEMBER : Time.[Hierarchy].Currentmember.Lag(9), [Measures].[Sales Amount])/10
When I use this and When my current member is D12 Then I am getting the value as 200/10 = 20 (from D12 to D3).
However ideally it should be 230/10Â =23 (from D12 to D1) and lag should exclude D6 and D7 as it is a weekend.
I also have a working day flag in my time dimensions which says 0 for non working days and 1 for working days.
Any working example of what a ragged hierarchy should look like for the unary and custom rollup calcs to work?
I have a parent-child hierarchy that works as expected but can't manage to get the same with a flattened ragged hierarchy. Parents disappear if any of the children aren't in the fact table (even though the children are set to '~' and the parent has its own custom rollup calc).
Do I need to replicate the unary and measure formula all the way to the end in the same way i do the member name/value (so i can set ignore if same as parent)? Setting unary to null seems to work when the key doesnt exist in the fact table but the default behaviour for null is '+' so i end up getting wrong results if the key does exist!
Repro:
In words
I have an EAV style fact table. The measure name is defined in a 'dim measure' table. Some measures exist in the fact, some are manufactured using Unary ops or custom rollup formulas.
Here's a diagram of the hierarchy
I want to recreate the functionality as a flattened hierarchy (the number of levels is fixed), but can't seem to do it... Here's what I've gotten for the same hierarchy created using the two different approaches
Below are queries to recreate the tables from adventureworksdw2014. The parent child hierarchy is set up the standard way... measureparentkey is parent. set name, unary and formula as expected. Set NonLeafDataHidden for the parent attribute. I've tried various combinations for the ragged hierarchy but none work 100% of the time. I want to have a user defined hierarchy for the performance benefits.
--Fact view
CREATE View dbo.FactSales
AS
select fis.ProductKey
, fis.OrderDateKey
, 1 AS MeasureKey -- salesAmount
, fis.SalesAmount AS MeasureValue
[Code] .....
I have this simplified query below to illustrate what I'm looking for. I would like to get the data for the time between the two dates in the where clause ('01-05-2015' and '10-03-2015'). In SQL it would be a BETWEEN. I'm looking for the MDX equivalent.
I have these dates hard coded but what I eventually will be doing is an @BeginDate / @EndDate parameter in an SSRS report.Â
SELECT [Measures].[Ship Resale S&D Run Rate] ON COLUMNS,
[Branches].[Region].[Region] ON ROWS
FROM [Sales]
WHERE [Time].[Date].&[01/05/2015]:[Time].[Date].&[10/30/2015]
*WHERE @BeginDate : @EndDate
I have a date dimension set in the SSAS Cube. I have been trying get quarter,year,month,semester start and end dates using ClosingPeriod() and OpeningPeriod() functions but not getting the exact value. How the get correct dates for a given date.
View 2 Replies View RelatedI've saw many tutorials about using TimeSeries. But all of them using a table. But I'm using a cube to represent data. So I'm trying to build forecast from cube, but it doesn't so good as in could be.
I've got the same problem as desribed in Microsoft's tutoral Adventure work. So I need to forecast a series of sales.
The problem is that I can't create second key value, as it shown in tutorial. So I can't split good's sales. I have created dimentions for goods and for time. So cube's browser shows me very handsome view, but the problem with mining model still remains...
Please, help me! How can I solve this problem?
Can I create a separate table from cube to build forecast by this table?
Or I can solve this problem not using tables?
Hi,
We are trying to forcast product sales for next three months based on their sales for previous 12 months. In this case, Microsoft Time Series algorithm requires the sales data to be present for each product for past 12 months (???). However, our products have typical life span of 6 months and obviously the new products will not have sales before they were added. Any help will be very much appreciated.
Thanks
Riju
In order to setup my forecasting mining model I have created a special view that runs against my fact table and creates time series on the level I need.
Code Snippet
Select DFUKEY, DATE, QTY from Dim_FACT where DFKEY like '020%'
So I get the following input fr my model:
time series key (e.g. DFUKEY)
date (time key)
QTY (to be predicted)
For testing purposes I created a small view (similar to AdventureWorks) that only contained a few time series. The model was created and processed in ~2 minutes or less. The viewer came up almost immediately and I was able to see results.
Now my real view has about 25000 time series I need a forecast for and that I also like to review in the viewer. If I create a mining model against that bigger view the processing takes ~15m or so and the viewer is likely to time out.
The worst part thought is when I try to get the forecast for a time series (see query below) it takes minutes before the answers come back.
Code Snippet
SELECT FLATTENED [DFUKEY], PredictTimeSeries([QTY], 12)
FROM [TimeSeries]
WHERE [DFUKEY] = '02078147AG'
I would understand if without the WHERE clause it would take long but as is?
I am pretty sure I created the model the right way so I wonder if somebody can give me recommendations as to how forecast larger numbers.
Finally I intend to call the foarcast for a particular series using an Action from within a cube.
Thanks in advance.
Dirk
Hi All,
Does anyone know how is forecasting defined in the Data Mining Add-In (Excel)? If we look to the web, we would see different Forecasting Standards. What specific Forecasting algorithm and standards used in the Forecasting in the Data Mining Add-In (Excel)?
Regards,
Joseph
I€™m having a problem with Excel 2007 DM and SQL 2005 and I hope someone out there has a solution.
Consider the following environment:
Windows XP SP2 or Windows Vista, Excel 2007, Data Mining Add-in, SSAS 2005 (with session mining models enabled, an AdventureWorksDW cube deployed and drill-through actions available).
Now take the following steps:
1. In Excel 2007 set up a connection to SSAS
2. Connect to the cube and create a new pivot table report (drag and drop whatever you like)
3. Right-click on one of the cell values in the data region and either select a drill-through action, or, select Show Details in the context menu
4. Ensure that you have at least 10 detailed records that are generated on a new worksheet page; you should have a time-based column in your detailed records
5. Select the table of detailed data, then select the Analyze tab (within the Table Tools grouping) which appears in the topmost menu above the ribbon
6. Click the Forecast button in the ribbon and choose both the field which you want to predict as well as the time-based column (from step 4) as well as the number of time periods to forecast
7. Finally click OK.
1. Having followed these steps on both WinXP SP2 and Vista, I keep coming across the exception: HResult:0x800A03EC. Any ideas as to why this exception pops up? If I was using a normal table of data (which was not generated from a Show Details or drill-through action), then the Forecast button works fine.
I googled it and thought the localization settings for SSAS 2005 and Excel 2007 needed to be the same (initially they weren€™t). I€™ve tried removing the auto-filters which appear atop each column in the detailed data table prior to clicking the Forecast button, and, I€™ve also tested for a series of data across a number of time periods with the same result.
Also, a colleague of mine discovered that the column headers that appear by default from a drill-through start with "$[", and, in removing them the Forecast function appears to work.
I would have thought there would be a seamless transition in Excel 2007 between data retrieved from a cube and the DM Add-in featueres (or at the very least, a more meaningful exception message than the one presented).
Is there something I€™ve missed, or, is there a KB article I haven€™t come across yet? As I know for a fact that the problem is reproducible, is there a fix to this problem on its way to us? Is there a useful workaround that doesn't require manual intervention?
Thanks,
A.
I have another question I'm just looking for some opinions on this. I created a database through sql (sql engine, access 2000 client side) now this database has Hyperlinks in it which lead to the employee's documents which on in acrobat pdf. Does anyone for see an problems that might occur wth that???
Thanks again
Hi,
We have a NT service built via Visual Basic 6 which fires dts processes on demand according some criteria. Well, on the whole and basically only for the following rule:
-dts process is executed when our service recognize a concrete plain file leaved in a folder. It€™s generally generated for a mainframe JCL along with a activator file (.nul file, empty). Each application own a folder and each folder have lots of files as the aforementioned ones. Scheduler for that is an ASP application that lives within a intranet. It also is a monitor and logger
In that schema, keeping on mind that eight out of ten are ETL processes, critical ones and on daily-basis, our question is, how to handle on with this stuff using by sql25k and its powerful SSIS?? Migrating such dts packages to SSIS packages is a piece of cake (now no before; thanks Jamie, Michael and other guys for your awesome help) because of its tipology. However our main concern is what the hell to use. Centralized dtsx package running all day long as a service, listening and checking for files? .Net windows application either vb or c#? aspx solution? Set of .vbs interlinked among them?
This is a old post/goal. In spite of we have not defined any deadline for that, what the heck as time goes by is more urgent for us. A restriction: we can€™t to programme nothing from our sql25k production cluster physically. Any stuff will be done on workstations.
In terms of cost, no problem at all. We are interested in hear thoughts and point of view about which is the most effective way
Thanks in advance,
How should I find the dates for 3 days past and 5 days future. Such as TodayDate-3 and TodayDate+5 base in the date only, discarding the time?
View 5 Replies View RelatedI have had one tossed at me that I've never dealt with before. While I've done calculations on date parts I'm a little lost on how to proceed. They want a stored procedure that will take date from one field and add three days to it and that would update another field. The three days must account for weekend and holidays.
The date would be receiveddate 12/03/2007 and the due date 12/06/2007. If someone has run into this and has some sample code it would be a great help.
Any help would be appreciated.
How should I find the dates for 3 days past and 5 days future. Such as TodayDate-3 and TodayDate+5 base in the date only, discarding the time?
View 20 Replies View RelatedHello,
I just create FLATFILE connection manager from UI, and now I would like save this connection manager for future use. I am not able to do this and now I am really stuck in the middle of coding.
ConnectionManager class has SaveToXML and LoadFromXML methods from IDTSPersist, these methods are not for use, but I am able to save CM to XML, however not load. Every try for loading ends with error 0xC0011008. The XmlNode.Name == "DTS:ConnectionManager"
CM class is not serializable, thus standard .NET feature is not working, also is sealed.
IDTSConnectionManagerFlatFile90 of cm.InnerObject is not serializable either, because it is interface and cm.InnerObject cannot be cast to ConnectionManagerFlatFileClass because is the COM.
Is there any way, how to save and load ConnectionManager besides writing of tons of code for each of connection manager type?
I am able to save and load package with connection manager, but it is not exactly what I want. Also is there possibility to work with XML in packages, but this is not "right".
Thanks
Erik
I need to test an HR application with a future date without changing the system date. If I just rollback the the date fields in the data, it may not work as the application itself might be using getdate() type of functions..
Any thoughts on how to do this w/o touching the system date setting?
Thanks.
Hi,
I'm trying to migrate to VS2005 and prepare for future Windows Mobile releases. My app uses MFC, CCeDBDatabase, and RAPI from the desktop to sync with the device version of the app. Works great now, but I'm worried about how many different ways it will be broken in the future .
My questions - first, are we pretty certain that CEDB will NOT be supported after WM5? Next, is there, or will there be, a CCeDBDatabase class for EDB? VS2005 seems to know nothing about CCeDBDatabase. What should I be doing with the CCeDBDatabase; do I need to re-write everything using only API calls? Also, there is no RAPI for EDB, is this correct? If so how can I switch to EDB and still communicate/sync with the desktop using RAPI? As always, any advice is greatly appreciated, thanks!
Mike B.
Can anyone comment on the future for spatial data types in SQL CE? This would be a very useful feature for mobile devices connected to a GPS.
View 4 Replies View RelatedBooks online "Deprecated Database Engine Features in SQL Server 2005" makes references to "Features not supported in Future SQL Server versions".
http://msdn2.microsoft.com/en-us/library/ms143729.aspx
Will there be an Upgrade Advisor available to help us identify issues with our code any time soon or should we start building our own?
According to SQL Server 2005 Book Online, the following "Important" note applies to SET_ANSI_NULLS and SET_ANSI_PADDING:
"This feature will be removed in a future version of Microsoft SQL Server. Avoid using this feature in new development work, and plan to modify applications that currently use this feature."
However, these options are automatically inserted in the auto-generated scripts (such as CREATE TABLE) in SQL Server 2005 Management Studio.
I am working on new scripts that include these options. What are the replacements for these options? Does the above note apply to ALTER DATABASE options? Please give me advice!
Thanks!
I have a sql server express 2005 database with a table named Events with a column named Date (datetime datatype). I want a query that will display all rows that are either current or future but not past. I suspect there is a simple way of doing this. As a Newbie searching archived threads this is what I have come up with so far. I determine the number of days from present:SELECT DATEDIFF(day, Date, GETDATE()) AS NumberOfDays FROM EventsThis yields number of days from present with positive numbers in the past and negative numbers in the future. Thus setting a WHERE clause to <= 0 would limit my results to present or future events. Something like this:SELECT * FROM EventsWhere DATEDIFF(day, Date, GETDATE()) AS NumberOfDays <= 0The error message states: "Incorrect syntax near the keyword 'AS'"This feels like a clumsy way to approach this problem, but I have to start where I am.Any suggestions on how to proceed will be greatly appreciated.
View 2 Replies View RelatedHello all!
I want to get into consulting (perhaps a little at first and more once I can support myself on it solely). I am about to start the purchase of a laptop soon, and I want to know which components I should focus on the most. I also want to know which version of SQL Server I should purchase.
From my studies and what I have learned (about to take the test to get my sql server 2005 certificate) I am going to assume that I want to get SQL Server 2005 developer software. However, the part that is not touched on very often, is if I am working with a client that has say, SQL Server 2005 enterprise, x64 exition, and I have developer IA64 edition, would I be able to connect and work on their server?
As far as the hardware is concerned, I am thinking I want a laptop with a large enough screen (I currently get a discount through hp where I currently work and am looking at the models that have a 17 inch screen) so I don't go blind trying to work . I am also thinking I don't need an extremely powerful processor, but I am not sure between athlon and pentium on which I Should choose since I do not have much experience in this area with sql server. I know I want at least 2GB of memory for the simple fact that sql server isnt the only thing I will be doing on the laptop and want some headroom. I also want a large enough hard drive to hold everything, but I definitely want to see if I can get a 7200rpm drive vs a 5400rpm for the simple fact that I have noticed a significant difference in read and write speeds between the 2 drives. I am not too fussed over the video card, as I am not into graphics design or high-end video games (I think world of warcraft is the most intense I have bought in years). I am also thinking that because I wont need to have a very powerful video card, that a mega powerful battery is not super important.
These so far are my thoughts. Please let me know your opinions on what I should focus on/get
I want to query my msdb job and jobschedule related tables to generate a list of runtimes for each of these jobs for the next day or any future date. This query should output JobID, Run_Date(YYYYMMDD), and Run_Time(HHMMSS).
If I have 3 jobs with...
Job#1 scheduled to run once every 4 hours between 6 AM and 10 PM
Job# 2 scheduled to run every 15 minutes between 11 AM and 1 PM
Job# 3 scheduled to run every minute between 4 PM and 4:15 PM
my output should look as below ....
Hi,
I've got a time series of the number of new customer subscriptions, which
is the target attribute to predict. The number of subscriptions depends on
various marketing activities, such as mailings, which are known within
the time series for the past.
If I train an ART (MS Autoregression Tree), it learns the trend pattern
as well as the correlations between the marketing activities and the
target (cross-correlations), right?
What I would like to do is, providing the model with some marketing activity
planning for the future and let the model predict the number of new
subscriptions based on a) the past trend pattern and b) the future activities.
Unfortunately a time series algorithm does not provide some kind of scoring for input data.
What would be the best approach to solve this problem? How about linear regression?
But how to train a regression model with trend patterns?
Thanks for your help!
i need to be able to do an sql call whereby i can check the day and hour for publishing.eg select story where day = ?? AND hour = ?? any one know the best way to do this.
i need to call out date as a day number or someting and the getdate as just the hour??? anyone know how??