How to color code non aggregated tabular data based on increase of measure value ?

[shekhar_analyti]

Hi All ,

How to color code non aggregated tabular data based on increase of measure value ?

Table

Measure	Value	Year	Country	Region	Item	Component	Brand
Rank	8	2010	India	North	Analgesic	BTB	Cipla
Rank	8	2011	Japan	Central	Sedative	Acetylamylin	Pfizer
Rank	7	2009	Japan	Central	Sedative	Acetylamylin	Pfizer
Rank	7	2009	Korea	Coastal	Sedative	Acetylamylin	Pfizer
Rank	7	2009	Japan	Bay	Sedative	Acetylamylin	Pfizer
Rank	6	2009	India	North	Analgesic	BTB	Pfizer
Hold	5	2010	Japan	Central	Sedative	Acetylamylin	Lupin
Hold	5	2010	Korea	Coastal	Sedative	Acetylamylin	Lupin
Hold	5	2010	Japan	Bay	Sedative	Acetylamylin	Lupin
Hold	4	2010	India	North	Analgesic	BTB	Lupin
Hold	4	2011	Korea	Coastal	Sedative	Acetylamylin	Lupin
Rejection_Index	4	2013	Japan	Bay	Sedative	Acetylamylin	Pfizer
Rejection_Index	3	2011	Japan	Central	Sedative	Acetylamylin	Pfizer
Rejection_Index	3	2012	Japan	Central	Sedative	Acetylamylin	Pfizer
Rejection_Index	2	2011	Japan	Central	Sedative	Acetylamylin	Pfizer
Rejection_Index	2	2011	Korea	Coastal	Sedative	Acetylamylin	Pfizer
Rejection_Index	2	2011	Japan	Bay	Sedative	Acetylamylin	Pfizer

OutPut

[shekhar_analyti]

I have attached sample data with expected output and logic

[shekhar_analyti]

[shekhar_analyti]

stalwar1‌ ... Anything wrong Sir , with my data

[sunny_talwar]

Sir?? ... Whatever happened to Sunny Bhai

Nothing wrong with your data... I just didn't open the Excel file until you mentioned that in your post

I have attached sample data with expected output and logic

[shekhar_analyti]

Accha .. i got it . I thought that by mistake i have uploaded wrong data , possibly a blank sheet .

Again ... needed your help Sunny Bhai .

Actually i had to show data with

1) Colour coding in Table (which is already there in excel)

2) List Box based selection :  If user select field value as Positive then aap show show only product those whose mesaure values are increasing and if selection is Negative then it should show measure values of product which got decreased over period of time .

Thanks & Regards

Shekar

[sunny_talwar · Accepted solution]

May be this

Legend:

LOAD AutoNumber(Measure&Country&Region&Item&Component&Brand) as Key,

Measure,

    Value,

    Year,

    Country,

    Region,

    Item,

    Component,

    Brand

FROM

[Colour on Aggr Data.xlsx]

(ooxml, embedded labels, table is Table);

FinalTable:

LOAD *,

If(Key = Previous(Key), If(Value > Previous(Value), 'Positive', If(Value < Previous(Value), 'Negative'))) as Trend

Resident Legend

Order By Key, Year;

DROP Table Legend;

[shekhar_analyti]

Thank You Bhai ...

I have just one question , that cant there be an alternative to above solution .

My only concern is that the Key you have created is from multiple field , and what if real time data is huge and full of unique values . Wont this key creation degrade the performance ?

I am sure this question might be very silly for a Legend   .

[sunny_talwar]

I have not done a lot of testing around AutoNumber, but I have found it to be slightly slower with large amount of data.... other alternative is this....

Legend:

LOAD Measure,

     Value,

     Year,

     Country,

     Region,

     Item,

     Component,

     Brand

FROM

[Colour on Aggr Data.xlsx]

(ooxml, embedded labels, table is Table);

FinalTable:

LOAD *,

If(Measure = Previous(Measure) and Country = Previous(Country) and Region = Previous(Region) and

Item = Previous(Item) and Component = Previous(Component) and Brand = Previous(Brand),

If(Value > Previous(Value), 'Positive',

If(Value < Previous(Value), 'Negative'))) as Trend

Resident Legend

Order By Measure, Country, Region, Item, Component, Brand, Year;

DROP Table Legend;

But any which way you would have to Order By all your fields which form your key and use them in the If statement....

[shekhar_analyti]

Sunny Bhai , I observed that for one Scenario , Trend Setting is not working .

If you see Rejection Index is coming negative for one scenario for year 2011 , but instead of that value negative should have populated for year 2012 ...
Actually year 2011 have two entry , in such scenarios that record with max value of  given measure should be used for comparison .

I have saved that in qvw for easy reference .

Note : I have made slight change it data .

Thanks & Regards

Shekar