Rule modification not working
Posted: Thu May 19, 2016 10:16 am
Hello:
I have an existing rule / feeder that I need to modify that I am having some difficulty with and was wondering if anyone could please help.
The logic behind this rule calculation is that it is:
For each and every broker (dim 'zz BDE') that has Gross Gross Written Premium in an Underwriting Year, it is a count of the number of products (this is dimension 'zz Class') that the Gross Gross Written Premium was written against.
The shape of the cube 'zz Commercial' that the rule is based off of is:
zz BDE
zz Broker
zz Class
zz Development Month
zz Underwriting Year
zz Accident Period
zz Policy Event Type
zz Version
zz Heads of Damage
zz Tenure
zz Measures
zz BDE and zz Broker are nearly identical dimensions - zz BDE has additional hierarchies but the data at the N level is the same.
zz Class is where the products are that I am counting. Currently the rule counts level one elements as the relevant products.
First problem is I want the data to show for dimension 'zz Development Months'.
The original rule is as follows:
And Feeder
I modified this to remove 'zz Development Month':'Not Applicable' in both rule and feeder. After this change, data is pulled into the 'zz Development Month' dimension, however at the highest level, the data approximately triples which I wouldn't expect (and at the other levels as well, the count is higher than it was initially.)
The other problem is with the product count. As stated it counts all level one products in the 'zz Class' dimension if the broker has GGWP written against it. Instead of this, is there a way to hard code specific product elements in zz Class? I actually need select elements that sit in both levels one and two. I have played around with this but the rule completely breaks.
Many thanks for any assistance in advance for either or both of these issues.
I have an existing rule / feeder that I need to modify that I am having some difficulty with and was wondering if anyone could please help.
The logic behind this rule calculation is that it is:
For each and every broker (dim 'zz BDE') that has Gross Gross Written Premium in an Underwriting Year, it is a count of the number of products (this is dimension 'zz Class') that the Gross Gross Written Premium was written against.
The shape of the cube 'zz Commercial' that the rule is based off of is:
zz BDE
zz Broker
zz Class
zz Development Month
zz Underwriting Year
zz Accident Period
zz Policy Event Type
zz Version
zz Heads of Damage
zz Tenure
zz Measures
zz BDE and zz Broker are nearly identical dimensions - zz BDE has additional hierarchies but the data at the N level is the same.
zz Class is where the products are that I am counting. Currently the rule counts level one elements as the relevant products.
First problem is I want the data to show for dimension 'zz Development Months'.
The original rule is as follows:
Code: Select all
['zz Development Month':'Not Applicable',
'zz Policy Event Type':'Unspecified',
'zz Tenure':'Unspecified',
'zz Heads of Damage':'Not Applicable',
'Product Density Workings'] = n:
max(0,
sign(['All Development Months','All Policy Events','All Tenure Types','All HOD','Gross Gross Written Premium'])
);
# take first parent, write to first child
['zz Development Month':'Not Applicable',
'zz Policy Event Type':'Unspecified',
'zz Tenure':'Unspecified',
'zz Heads of Damage':'Not Applicable',
'Product Density'] = n:
sign(
DB('zz Commercial',!zz BDE,!zz Broker,
elpar('zz Class',!zz Class,1),
!zz Development Month, !zz Underwriting Year,!zz Accident Period,!zz Policy Event Type,!zz Version,!zz Heads of Damage,!zz Tenure,
'Product Density Workings')
) *
if(elcomp('zz Class',elpar('zz Class',!zz Class,1),1) @= !zz Class,
1,
#else
0);
Code: Select all
['All Development Months',
'All Tenure Types',
'All Policy Events',
'All HOD',
'Gross Gross Written Premium'] =>
['zz Development Month':'Not Applicable',
'zz Policy Event Type':'Unspecified',
'zz Tenure':'Unspecified',
'zz Heads of Damage':'Not Applicable',
'Product Density Workings'];
# feed workings to first child of parent
['zz Development Month':'Not Applicable',
'zz Policy Event Type':'Unspecified',
'zz Tenure':'Unspecified',
'zz Heads of Damage':'Not Applicable',
# feed to first child of parent
'Product Density Workings']=>
DB('zz Commercial',!zz BDE,!zz Broker,
elcomp('zz Class',elpar('zz Class',!zz Class,1),1),
!zz Development Month,!zz Underwriting Year,!zz Accident Period,!zz Policy Event Type,!zz Version,!zz Heads of Damage,!zz Tenure,'Product Density')
;The other problem is with the product count. As stated it counts all level one products in the 'zz Class' dimension if the broker has GGWP written against it. Instead of this, is there a way to hard code specific product elements in zz Class? I actually need select elements that sit in both levels one and two. I have played around with this but the rule completely breaks.
Many thanks for any assistance in advance for either or both of these issues.