/* line 1*/		filename gradepop
			'd:\ginger\econ424\econ424-fall2001\final\gradepop.csv';
/* line 2*/       filename player
             	'd:\ginger\econ424\econ424-fall2001\final\player-race.csv';

/* line 3 */      data gradepop;
/* line 4 */      infile gradepop firstobs=3 dlm=',' dsd missover;
/* line 5 */	informat printer $16. cardnum $4. playernm $50.;
/* line 6 */      input       year
                              printer
                              cardnum
                              playernm
                              n1
                              xbar1
                              sigma1
					n2
					xbar2
					sigma2;
/* line 7 */      growth=(n2-n1)/n1;
/* line 8 */      if year<=1979 then vintage=1; else vintage=0;
/* line 9 */      if year=. then delete; 
/* line 10 */     run;

/* line 11 */     proc contents; 
/* line 12 */     run;

/* line 13 */     proc freq data=gradepop;
/* line 14 */     tables vintage;
/* line 15 */     run;

/* line 16 */     proc means data=gradepop noprint;
/* line 17 */     class playernm;
/* line 18 */     var n1 n2;
/* line 19 */     output out=cardsum sum=sum_n1 sum_n2;
/* line 20 */     run;

/* line 21 */     proc sort data=cardsum; 
/* line 22 */     by sum_n1; 
/* line 23 */     run;

/* line 24 */     proc print data=cardsum; 
/* line 25 */     run;

/* line 26 */     proc sort data=cardsum; 
/* line 27 */     by sum_n2; 
/* line 28 */     run;

/* line 29 */     proc print data=cardsum; 
/* line 30 */     run;

/* line 31 */     data gradepop;     
/* line 32 */     set gradepop;
/* line 33 */     avgtest=(xbar1-xbar2)/sqrt(sigma1*sigma1/n1+sigma2*sigma2/n2);
/* line 34 */     if avgtest>=1.64485 then reject=1; else reject=0;
/* line 35 */     run;

            	/* What is 1.64485?
			Suppose a is a random variable that conforms to standard 				normal distribution, probability (a>=1.64485)=0.05 */

/* line 36 */     proc means data=gradepop;
/* line 37 */     class vintage;
/* line 38 */     var avgtest reject; run;

/* line 39 */     data player;
/* line 40 */     infile player firstobs=2 dlm=',' dsd;
/* line 41 */     informat playernm $50.;
/* line 42 */     input  race nopitch playernm; 
/* line 43 */     run;

/* line 44 */     proc sort data=gradepop; 
/* line 45 */     by playernm; 
/* line 46 */     run;
/* line 47 */     proc sort data=player; 
/* line 48 */     by playernm; 
/* line 49 */     run;

/* line 50 */     data merged;
/* line 51 */     merge gradepop (in=one)  player (in=two);
/* line 52 */     by playernm;
/* line 53 */     if vintage=1 & race=1
       			then raceage='vintage+white';
/* line 54 */     else if vintage=1 & race=0
				then raceage='vintage+non-white';
/* line 55 */     else if vintage=0 & race=1
				then raceage='modern + white';
/* line 56 */     else if vintage=0 & race=0
				then raceage='modern + non-white';

/* line 57 */     proc glm data=merged;
/* line 58 */     class raceage;
/* line 59 */     model growth=raceage;
/* line 60 */     mean raceage / waller;
/* line 61 */     mean raceage / lsd;
/* line 62 */     run;

/* line 63 */     proc reg data=merged;
/* line 64 */     model growth=race;
/* line 65 */     run;

/* line 66 */     proc reg data=merged;
/* line 67 */     model growth=race vintage;
/* line 68 */     run;


/* line 69 */     proc sort data=merged; by vintage; run;
/* line 70 */     proc reg data=merged;
/* line 71 */     by vintage;
/* line 72 */     model growth=race;
/* line 73 */     run;