HOW TO AGGREGATE WITH MISSING DATA

[IE]

Good morning,

I need one help for the example I attach.

I would like to show in one chart the comparation  between production and forecast.

It means data in every day with production and forecast.

The problem is in production there is not production every day and the chart lost the data.

What I would like is to show all days and the forecast allways must be 120 in all cases.

Thanks in advance

[deepanshuSh]

Its not making much sense to me, if you already have the forecast value set to 120, then it would alreadybe fixed right? then what are you trying to achieve here.

[IE]

Sorry.

The forecast can change, it is for simplificating the example.

Any idea?

[marcus_sommer]

It should work if production and forecast are in the same table - just concatenated.

- Marcus

[IE]

I prefer not to change nothing in the relation of the tables because can affect to other calculs.

The original Qlik it is a very complex .

The example is a litle summary of it.

[marcus_sommer]

The officially recommended data-model within Qlik is a star-scheme - means a single fact-table with all measures and n dimension-tables.

Everything else is possible but usually much more complex to develop and to maintain and quite often with serious disadvantages in regard to the performance, the relation-ship between the data and the usability - at least against the star-scheme model. Therefore the suggestion to consider a change of the data-model.

In regard to your missing data - they couldn't be linked between two tables without the lost of information. AFAIK there is no other possibility as to populate the missing data or the above mentioned simple way of concatenating them.

- Marcus 

[oskartoivonen]

Sounds like the missing production data is functioning as the link between your forecast data and for example calendar table. While the real solution would be to refactor into a star-scheme like @marcus_sommer suggested to allow forecast measure to function independently, another workaround is to add dummy production data with just keys and no values, that'll fill in the blanks and link the tables back up with minimal effect on existing calculations.