Reading in a reference table to perform an analysis on a data set and output csv files

[BenBrady]

I have a reference table that displays how my industries are grouped into sectors , and what has to go in each of the charts. The reference table looks as such:

Cluster	Sector	Industry1	Code1	industry2	code2	Table_num_per_sector	chart_title
RW	CP			Pharmacy Sales	6124	1	test1
A	AG	Ag nfd	010			4	test8
A	AG	Mush	012			4	test8
A	AG	Fruit	013			4	test8

I want to reference this table and output csv files. The data set I will perform the analysis on is this: 

code	year	industry	emp
010	2008	Ag nfd	1
010	2009	Ag nfd	2
4	2008	B	2
4	2009	B	2
012	2008	Mush	1
012	2009	Mush	2
013	2009	Fruit	3
6124	2008	Pharmacy Sales	2
6124	2009	Pharmacy Sales	4

I want to use the reference table to go through this data set and output csv files like this (for example).

CSV1: Called test1

year	Pharmacy Sales
2008	2
2009	4

CSV2: called test8

year	Ag nfd	Mush	Fruit
2008	1	1	0
2009	2	2	3

It is important to note how the outputed csv's are called what they are told to from the reference table, and that missing values such as 2009 for Fruit are given a value of 0. Code1,Code 2 and industry1,industry2  in the reference table are to be compared with code and industry in the dataset that will be analysed.

I am wandering how to go about doing this. As I am working on a data set similar to this but on a bigger scale.

[ballardw]

You might need to supply a little more detail on exactly HOW your reference table is used. Since the "analysis" you show is different then that tells me that the rows of the reference are used differently. Without a clue as to what may be wanted how to use your reference table is not obvious.

Call execute can be used with a data set to conditionally generate code, if you know which code you want. If you do not know which code you want then that is an entirely different set of questions.

We would need to know the purpose and possibly more values for each of the variables in the reference data set and then the type of "analysis" for each csv, especially if it is anything more than selecting records from the data set.