Slowly Changing Dimension (SCD) - Model Breakdown

[Anonymous]

I have a model where i have a slowly changing dimension called CATEGORY which takes the following changing values;

1. LOCALLY MANUFACTURED
2. SECTION 75 IMPORTS
3. NORMAL IMPORTS
4. IMPORT - LOCAL

A product's CATEGORY can change from say SECTION 75 IMPORTS to NORMAL IMPORTS, or from NORMAL IMPORTS to IMPORT - LOCAL etc.

I initially had the model without Slowly Changing Dimension but remodelled it with SCD as per the attached qvw document.  The model is consistent for the months of January, February, March and April with the % Value and actual values of the above categories staying in a consistent range.  However, for the month of May, all values are very low when compared to the historical values.

I then remodel the application with no slowly changing dimension and the values of  May become consistent with those of Jan, Feb, Mar and Apr as follows:

1. % Value of Section 75 Imports: 4.28% and actual value at $43 661.15
2. % Locally manufactured 10.7% and value $109 007.42
3. % Value of normal imports 70.9% and value $ 723 026.57

Where could i be going wrong with my model?  Your assistance will be greatly appreciated.

[adamdavi3s]

Hi,

I just wanted to check this is dummy data and not sensitive as it looks quite full on....

Can you give us an idea of what values you are expecting? Lets focus on one particular value for now, break down the expression and the data model and see whats going wrong

[Anonymous]

Thanks Adam

This is real data.  The expected values are as follows:

1. % Value of Section 75 Imports: 4.28% and actual value at $43 661.15
2. % Locally manufactured 10.7% and value $109 007.42
3. % Value of normal imports 70.9% and value $ 723 026.57

[Anonymous]

Managed to solve my problem