/* EXERCICE 1 */

/* question 1 */
data Mercury;
infile "Mercury.txt" expandtabs ;
 input id lake$ alkalinity ph calcium cholophyll avg_m nosamples minn maxx yrstd age_data ;
run;
/* question 2 */
proc print data=mercury;
run;    /* 53 obs */
/* question 3 */
data mercury_age;
  set mercury;
  if age_data=1;
  drop age_data;  /* 53 obs et 11 var */
  data no_age;
  set mercury;
  if age_data=0;
   drop age_data; /* 10 obs et 11 var */
  proc print data=mercury_age; run;
  proc print data=no_age; run;
  /* question 4 */
  proc means data=mercury_age;
  var alkalinity ph calcium cholophyll avg_m nosamples minn maxx yrstd; 
run; /* pas de données manquantes */
 proc corr data=mercury_age;
 var alkalinity ph avg_m yrstd; run; /* toutes les variables sont corrélées entre elles */
 /* question 5 */
 proc univariate data=mercury_age normal;
 var alkalinity yrstd; 
   histogram alkalinity yrstd/normal; 
  run; 
   /* question 6 */
  data n_mercury_age;
   set mercury_age;
   log_alkalinity=log(alkalinity);
   log_yrstd=log(yrstd);
   run;
   /* question 7 */
   proc univariate data=n_mercury_age normal;
 var log_alkalinity log_yrstd;   /* toutes les deux sont de loi Normale */
   histogram log_alkalinity log_yrstd/normal; 
  run; 
  /* question 8  */
proc means data=n_mercury_age;
  var log_alkalinity log_yrstd;
  run;   /* 2.99 et -0.82 */
  /* question 9 */
  data n2_mercury_age;
     set n_mercury_age;
	 diff=log_alkalinity-log_yrstd; 
	  proc univariate data=n2_mercury_age;
	    var diff;
		run;

		/* EXERCICE 2 */
proc iml;
	/* question 1  */
	use mercury;
	read all var {alkalinity ph avg_m yrstd} into X;
	close pharynx;
  print(X);
/* question 2 */
  v=x[,3]-x[,4];
print(v);
/* question 3*/
z=j(nrow(x),1,1);
do i=1 to nrow(z);
  if v[i]<0 then z[i]=-1;
  if v[i]=0 then v[i]=0;
  if v[i]>0 then v[i]=1;
end;
vz=v || z;
print(vz);
/* question 4  */
Y=X || vz;
print(Y);
/*  question 5 */
create nouveau from Y ;
append from Y;
close nouveau;
quit;

/* Exercice 3 */ 
data final;
merge mercury nouveau;
 v=col5; 
 z=col6;
 keep id lake alkalinity ph calcium cholophyll avg_m nosamples minn maxx yrstd age_data v z;
proc print data=final ; run;

/* Exercice 4 */
proc sort data=final ;
	    by pH; run;
		axis1 value=(c=green) label=(c=green "alkalinity");
		axis2 value=(c=black) label=(c=black "pH");
	  proc gplot data=final;
	  plot alkalinity*pH / vaxis=axis1 haxis=axis2;
	  symbol1 interpol=j color=blue v=star;
	  title "alkalinity fonction du pH";
	  run;
quit;

/*  Exercice 5 ***/
	  /* question 1 */
	  option mprint mlogic symbolgen;
	  %let tableau=final;
	  %let V1=pH;
	  %let V2=Alkalinity;
	  %let V3=Calcium;
	  %let V4=avg_m;
	  /***  question 2  **/
	  %macro regression(X1,X2,X3);
	  proc reg data=&tableau;
	    model &X3=&X1 &X2;
		run;
	  %mend regression;

	  
	  %regression(&V1,&V3,&V4)
	  %regression(&V2,&V1,&V4)