STAT660-01_f18-team-1_project1_data_analysis_by_LL.sas

*******************************************************************************;
**************** 80-character banner for column width reference ***************;
* (set window width to banner width to calibrate line length to 80 characters *;
*******************************************************************************;

*
This file uses the following analytic dataset to address several research
questions regarding FIFA 18 user information

Dataset Name: FIFA_Player_Data created in external file
STAT660-01_f18-team-1_project1_data_preparation.sas, which is assumed to be
in the same directory as this file

See included file for dataset properties
;

* environmental setup;

* set relative file import path to current directory (using standard SAS trick;

X "cd ""%substr(%sysget(SAS_EXECFILEPATH),1,%eval(%length(%sysget(SAS_EXECFILEPATH))-%length(%sysget(SAS_EXECFILENAME))))"""
;

* load external file that generates analytic dataset fifa18_analytic_file;
%include '.\STAT660-01_f18-team-1_project1_data_preparation.sas'
;

title1 
'Research Question: What is the age distribution amongst each club in FIFA 18?'
;

title2 
'Rationale: This could help identify which countries tend to focus on creating their players from youth as opposed to those who purchase their players'
;

footnote1
'seems that most leagues with a higher age distribution are amongst asian, usa, and english leagues while a  much younger age distribution is amongst latin american teams and african teams'
;

footnote2
'This is important to understand why some leagues are of higher caliber as well as why some countries perform better in international tournaments'
;

footnote3
'I believe this is the case since some countries are behind infrastructure wise and prefer to buy the star players then to focus on youth academies'
;

*
Methodology: I think we could take the proc means of the age variable and base
it off the clubs and even possibly get a by league by club break down to 
understand the age distribution amongst top clubs and top leagues


Limitations: I think our analysis won't account for which players are 
designated starters on the team and which players are reserves
 
Possible Follow Up: Maybe a variable for starting position on the team whether 
it is starter, bench, or reserves
;

proc means 
		data=fifa18_analytic_file
	; 
   	class club
	;
	var age
	;
run;
title;
footnote;

title1 
'Research Question: What is the Euro Value distribution amongst each league and nationality in FIFA 18?'
; 
title2 
'Rationale: This could correlate based off the first question determining which teams "buy their trophies" as opposed to work for them'
;

footnote1
'based on the above tables we can see that the german, african, and english leagues favor a bigger body type then latin teams who prefer a smaller body type'
;

footnote2
'This is important to determine the playstyles and differences in team compositions'
;

footnote3
'This might be due to the fact that some teams are much more dominant with defense and aggressive plays as opposed to dribbling and finesse'
;


*
Methodology: Could proc sort body type by leagues and then find the mean salary
of players based off body types to find any correlation amongst leagues that 
might prefer more finesse players versus physical players

Limitations: I can see how the data might not tell us much as I am sure player 
position might play a role on what body type a club would want that player to 
have.
 
Possible Follow Up: Perhaps we could do the focus more on weight and height in 
order to determine whether a league prefers a faster/lighter finesse player or 
a physically bigger and muscular "power" player 
;

ods graphics on / width = 8in height = 8in;
proc sgplot 
		data=fifa18_analytic_file
	; 
  	vbox
		eur_wage / category = nationality
	;
	where overall > 80
	;
run;
title;
footnote;

title1 
'Research Question: How does the body type distribution differ in each league and is there a correlation with salary based on the preferences of each league?'
; 
title2 
'Rationale: This would help determine if leagues prefer different types of player body types based on league playstyles'
;

footnote1
'body type does not necessarily correlate with value, might want to check and see if we can tie this more over to skill level or game statistics as body type does not associate over with skill level'
;

footnote2
'This is important because even leagues that preferred bigger body types did not correlate with the pay of the body type showing that skill is valued higher'
;

footnote3
'not sure why this is the case, if your team is centered around a specific playstyle, why throw in a player that is the complete opposite?'
;

*
Methodology: Could we take the median and range of salaries based off of clubs,
leagues, and nationality, to see if there is a possible trend in the types of 
players that clubs spend money on (i.e. Brazilian) as well as the top clubs 
versus the smaller clubs

Limitations: I think the question or data might be too broad, would need to 
identify the best way to determine which leagues and/or clubs buy their 
trophies and players as opposed to raising their youth to play in their clubs 
 
Possible Follow Up: Perhaps we could determine the distribution of foreign 
players in each league and then take the mean salaries and values of each 
leauge based on that percentage of non-home grown players
;

proc corr 
		data=fifa18_analytic_file
	; 
  	var
		height_cm weight_kg
	;
	with
		eur_value
	;	
run;
title;
footnote;