W3.txt

LIBNAME ACX3300 'C:\Users\15758\Desktop\data3300\W3';
_________________________________________
* SAS Functions;
* Arithmetic/Mathematical;
/*
ABS(X)            ABS(-9)=9
EXP(X)            EXP(1)=e=2.71828
LOG(X)            LOGe(10)=2.30259
LOG10(X)        LOG10(10)=1
FACT(X)           FACT(7)=7x6x5x4x3x2x1=5040
MAX(X)
MEAN(X)
MEDIAN(X)
MIN(X)
MOD(X,K)        remainder x/k, MOD(5,3)=2, MOD(6,2)=0
N(x1,x2,x3...)    # of non-missing values, N(1,2,.,4)=3
NMISS(x1,x2,x3...)    # of missing values, N(1,2,.,4)=1
ORDINAL(k,x1,x2,x3...)  find kth num in sorted list ORDINAL(1,11,6,9,3)=3
RANUNI(0)      random number
RANUNI(X)      uniform/same random number
RANNOR(seed) random number from normal distribution with mean 0 sd 1
SIGN(X)           return sign of X as 1 or -1
SQRT(X)           SQRT(9)=3
SUM(X)            SUM(x1,x2,x3)  SUM(OF x1-x3) ignore missing value
                         SUM(x1-x3)  keep missing value=.
*/
/*
ROUND (3.1415, .01) *Returns the value 3.14;
ROUND(107,5) * Returns the value 105;
ROUND(3.6234)* Returns the value 4 (rounds to integer);

                       -3.5  -2.9  -1.1  1.1  2.9  3.5
ROUND          -4     -3     -1     1     3     4
CEIL                -3     -2     -1     2     3     4
FLOUR            -4     -3     -2     1     2     3      
INT                 -3      -2     -1    1     2     3
INT+SIGN      -4       -3    -2     2     3    4
*/
-----------------------------------------
* Date and Time;
DATA DATE ; 
DATE = MDY(03, 16, 2021);
YEAR = YEAR(DATE) ;
MONTH = MONTH(DATE);
DAY = DAY(DATE);
QUARTER = QTR(DATE);
WEEK = WEEK(DATE) ;
WEEKDAY = WEEKDAY(DATE);
PROC PRINT ; RUN ;

* dates interval;
DATA DATES;
INPUT @1 BDATE MMDDYY8.;
TARGET=TODAY(); 
AGE=INTCK('YEAR',BDATE,TARGET); 
DATALINES;
07101952
07041776
01011900
;
PROC PRINT DATA=DATES;
FORMAT BDATE WEEKDATE. TARGET DATE9.;
RUN;

/*vaccines are available today and everyday 25,000 people can get vaccinated. By the end of the year 2021, how many people will be vaccinated*/
DATA ONE;
PEOPLE = INTCK("DAY",TODAY(), MDY(12,31,2021))*25000;
RUN;
PROC PRINT; RUN;
-----------------------------------------
* Character;
/*
UPCASE		makes all the letters upper case
LOWCASE	makes all the letters lower case
TRIM 		removes trailing blanks
STRIP		removes all the leading and trailing blanks
LEFT		left aligns a character string
RIGHT		right aligns a character string
CHAR		returns the character in a specific position of a string
CAT		Connects character variables
*/

* SUBSTR;
DATA STRING ; A ="ABCDEFG";
B=SUBSTR(A, 2, 3);
PROC PRINT; RUN ;
/*B="BCD"*/
DATA STRING ; A ="ABCDEFG";
SUBSTR(A, 1, 3)="123";
PROC PRINT; RUN ;
/*A="123DEFG"*/
-----------------------------------------
* Access Previous Observations (Lags);
/*RET=(PRICE in day t – PRICE in day t-1)/Price in day t-1 = Pt/Pt-1 – 1*/
PROC SORT DATA=SASHELP.STOCKS OUT=STOCKS; BY STOCK DESCENDING DATE ; 
DATA STOCKS1; SET STOCKS ; BY STOCK DESCENDING DATE ;
NEXT_P=LAG(ADJCLOSE);
RET=LAG(ADJCLOSE)/ADJCLOSE – 1 ;
IF FIRST.STOCK THEN DO ;
NEXT_P=. ; RET= . ; 
END ;
RUN ;
DATA CHECK; SET STOCKS1; BY STOCK DESCENDING DATE ;
IF FIRST.STOCK ; 
PROC PRINT ; RUN ;

* Calculate the stock returns in each year;
DATA STOCKS; SET SASHELP.STOCKS (KEEP=DATE STOCK ADJCLOSE);
WHERE MONTH(DATE)=1;
YEAR=YEAR(DATE);
PROC SORT DATA=STOCKS; BY STOCK DESCENDING DATE; RUN;
DATA STOCKS1; SET STOCKS; BY STOCK DESCENDING DATE;
NEXT_P=LAG(ADJCLOSE);
RET=LAG(ADJCLOSE)/ADJCLOSE-1;
IF FIRST.STOCK THEN DO;
NEXT_P=.; RET=.;
END;
RUN;
PROC PRINT DATA=STOCKS1(OBS=20); RUN;

* Calculate the stock returns for the period from 1990 - 2005;
DATA STOCKS2; SET SASHELP.STOCKS (KEEP=DATE STOCK ADJCLOSE);
WHERE MONTH(DATE)=1 AND YEAR(DATE) IN (1990,2005);
PROC SORT DATA=STOCKS2; BY STOCK DESCENDING DATE; RUN;
DATA STOCKS3; SET STOCKS2; BY STOCK DESCENDING DATE;
NEXT_P=LAG(ADJCLOSE);
RET=LAG(ADJCLOSE)/ADJCLOSE-1;
IF FIRST.STOCK THEN DO;
NEXT_P=.; RET=.;
END;
RUN;
PROC PRINT DATA=STOCKS3(OBS=20); RUN;

* Which company has the highest stock returns in the period from 2000 -2005;

/*
not easy to find uniform begin and end date - earnings announcement
1 select dataset containing daily/weekly/monthly returns during period 
P1-P4; RET1=P2/P1-1, RET2=P3/P2-1, RET3=P4/P3-1 
total returns = P4/P1-1
2 take log of (RET+1) then sum up
log(RET1+1)=log(P2/P1)=log(P2)-log(P1)
SUM=sum(log[(RET1-3) +1])=log(P4/P1)
3 EXP function to transform log returns into raw return
EXP(SUM)=P4/P1
*/
PROC SORT DATA=SASHELP.STOCKS OUT=STOCKS; BY STOCK DESCENDING DATE;
DATA STOCKS1; SET STOCKS; BY STOCK DESCENDING DATE;
RET=LAG(ADJCLOSE)/ADJCLOSE-1;
IF FIRST.STOCK THEN RET=.;
YEAR=YEAR(DATE);
RUN;
PROC SQL;
CREATE TABLE STOCKS2 AS
SELECT DISTINCT STOCK, YEAR, EXP(SUM(LOG(1+RET)))-1AS RET, COUNT(RET) AS MONTHS
FROM STOCKS1
GROUP BY STOCK, YEAR;
QUIT;
PROC PRINT DATA=STOCKS2(OBS=20); RUN;
_________________________________________
* SAS ARRAY - simplified;
/*
ARRAY TIME(1:6) A B C D E F ; * creates six variables TIME(1)=A, TIME(2)=B, etc ;
ARRAY NUM(5) 1, 2, 3, 4, 5;  * creates five variables NUM(1)=1, NUM(2)=2, etc;
ARRAY A (*) $ Q1-Q50; * create a number of variables A(1)=Q1, A(2)=Q2, etc;
ARRAY CUPSIZE [*] CUP1-CUP4 (8,16,24,32); create a set of variables CUPSIZE(1)=CUP1=8, CUPSIZE(2)=CUP2=16, etc;
ARRAY SIC(100) 1-100;  * creates 100 variables SIC(1)=1, SIC(2)=2, etc;
*/
DATA A;
FORMAT ONE TWO THREE FOUR $10.;
INPUT ONE $ TWO $ THREE $ FOUR $;
DATALINES;
APPLE ORANGE PINEAPPLE APRICOT
LEMON APPLE KIWI STRAWBERRY
;
RUN;
DATA B; SET A;
ARRAY FRUIT(4) ONE TWO THREE FOUR;
IF "ORANGE" IN FRUIT THEN ISORANGE=1;ELSE ISORANGE=0;
PROC PRINT DATA=B;
RUN;

* vs complex code;
DATA B; SET A;
IF "ORANGE" IN ONE THEN ISORANGE=1;
ELSE IF "ORANGE" IN TWO THEN ISORANGE=1;
ELSE IF "ORANGE" IN THREE THEN ISORANGE=1;
ELSE IF "ORANGE" IN FOUR THEN ISORANGE=1;
ELSE ISORANGE=0;
RUN;
-----------------------------------------
* DO LOOP;
/*
IF and DO: Use a DO in an IF statement to cause a group of statements to be conditionally executed.
Iterative DO loop: Iterate through a series of statements based on some incremental index.
DO UNTIL: Execute a group of statements until some condition is met.
DO WHILE: Execute a group of statements as long as specified condition is true.
*/

* Iterative DO loop;
SUM=0;
DO K=1 to 4;
SUM=SUM + K;
END; 

* DO UNTIL, put the condition to stop the iterations ;
K=1
SUM=0;
DO UNTIL (K GT 4);
SUM=SUM+K;
K+1 ;
END;

* DO WHILE, the loop continues until the condition is no longer true;
K=l;
SUM=0;
DO WHILE(K LT 5);
SUM=SUM+K;
K+l;
END;

* DO LOOP + ARRAYS;
/* if any of these readings is above 140, you want to classify the subject with the value HIGHBP=1, or else the subject will have the value HIGHBP=0*/
DATA PRESSURE;SET MEDDATA;
ARRAY SBP(5) READINGl-READING5;
HIGHBP=0;
DO I=l TO 5;
IF SBP(I) GT 140 THEN HIGHBP=l;
END;

/* If any of the prices is higher than the unit price (Price) then we define HIGHPRICE=1, otherwise HIGHPRICE=0
What if we don’t know how the number of products/prices*/
DATA PRICE;SET SASHELP.PRICEDATA;
TOTAL=0;
ARRAY P(*) DISCOUNT--PRICE17;
DO I= LBOUND(P) to HBOUND(P);
TOTAL=SUM(of TOTAL,P(i));
END;
DROP I;
RUN;
-----------------------------------------
* RETAIN statement;
* initial value of 1;
RETAIN SCORE 1;
* initial value of NONE & initial values of 0;
RETAIN CATEGORY "NONE" TIME1-TIME10 0;
* set the values of a list;
RETAIN TIME1-TIME4 (1 2 3 4);
RETAIN TIME1-TIME4 (1, 2, 3, 4);
RETAIN TIME1-TIME4 (1:4);

* calculate the values from the first observation to current observation;
DATA DAYS;SET MYDATA;
IF _N_=1 THEN FIRST=VISIT_DATE;
RETAIN FIRST;
DAYS=VISIT_DATE-FIRST;
RUN;

* accumulate TOTAL, and to find the MAX value;
DATA ONE;
INPUT DAY SALES;
DATALINES;
1 23.40
2 32.50
 …
;
RUN;
DATA TWO; SET ONE;
IF _N_=1 THEN TOTAL=0; 
RETAIN TOTAL MAX;
TOTAL=SALES+TOTAL;
IF SALES>MAX then MAX=SALES;
RUN;

* Using dataset SASHELP.PRICEDATA to find;
* The total unit sales in each year for each product;
* The highest unit price in each year for each product;
* The total revenue in each year for each product;
DATA PRICE; SET SASHELP.PRICEDATA;
YEAR=YEAR(DATE);
KEEP PRODUCT YEAR DATE PRICE SALE;
RUN;
PROC SORT DATA=PRICE; BY PRODUCT YEAR DATE;
DATA PRICE1; SET PRICE; BY PRODUCT YEAR DATE;
IF FIRST.YEAR THEN DO;
TOTAL=0; REVENUE=0;
END; 
RETAIN TOTAL REVENUE MAX;
TOTAL=TOTAL+SALE;
REVENUE=REVENUE+SALE*PRICE;
IF PRICE>MAX THEN MAX=PRICE;
IF LAST.YEAR;
PROC PRINT DATA=PRICE1(OBS=20); RUN ;
_________________________________________
* MACRO;
%LET YEAR=2015;
IND_DAY=MDY(7,4,&YEAR);
%LET DEPT=ENGINEERING;
GROUP = "&DEPT"; 

* combine macro variables;
%LET PATH=C:\SASDATA\;
%LET DSN=SOMEDATA;
%LET CLASS=GP;
%LET SELECTVAR=AGE TIME1-TIME4;
PROC MEANS DATA="&PATH&DSN" MAXDEC=2;
CLASS &CLASS ;
VAR &SELECTVAR;
RUN;

* ambiguity;
%LET STATUS=PRE;
&STATUS. PRODUCTION;
-----------------------------------------
* SAS MACRO;
%MACRO REPORT(SUBJ, DSN);
DATA REPORT;SET "&DSN";
IF SUBJ=&SUBJ;
TITLE "REPORT ON SUBJECT# &SUBJ";
PROC PRINT NOOBS DATA=REPORT;
VAR GENDER TIME EXPOSED DIAGNOSED;
RUN;
%MEND REPORT;
%REPORT(SUBJ=001, DSN=C:\SASDATA\SUBJECTS);

* macro available for use in a program;
%INCLUDE "C:\SASDATA\DMACRODISCLAIM.SAS";

* %DO LOOP
%do i = 1 %to 5;
data gp&i;
set mysaslib.somedata;
where status=&i ;
run;
%end;

* dataset sashelp.pricedata;
* The total number of units sold in a year, The total revenue (in dollars) in a year;
* when one enters the product number and the year;
* the code should output the following information: PRODUCT, YEAR, TOTAL, REVENUE;
%MACRO PRODUCT (PRODUCT=, YEAR=);
PROC SQL ;
SELECT DISTINCT PRODUCT, YEAR(DATE) AS YEAR, SUM(SALE) AS TOTAL, SUM(SALE*PRICE) AS REVENUE
FROM SASHELP.PRICEDATA
WHERE PRODUCT=&PRODUCT AND YEAR(DATE)=&YEAR;
QUIT;
%MEND ;
%PRODUCT (PRODUCT=5, YEAR=2000); RUN;
%PRODUCT (PRODUCT=12, YEAR=1998); RUN; 
_________________________________________
* tutorial;
* Calculate the stock returns in the 12 month period from July to June;
DATA STOCKS; SET SASHELP.STOCKS (KEEP=STOCK DATE ADJCLOSE); BY STOCK;
RET=LAG(ADJCLOSE)/ADJCLOSE-1 ;
IF FIRST.STOCK THEN RET=.;
IF MONTH(DATE)<7 THEN YEAR=YEAR(DATE)-1;
ELSE YEAR=YEAR(DATE);
RUN;
PROC SQL;
CREATE TABLE STOCKS1 AS 
SELECT STOCK, YEAR, EXP(SUM(LOG(1+RET)))-1 AS ARET, COUNT(RET) AS N_MONTHS
FROM STOCKS
GROUP BY STOCK, YEAR;
QUIT;

/*Calculate the stock returns in the three months 
after the earnings announcements. That is, for IBM, stock returns in March - May. Then calculate the average 
post-announcement returns across the years. */
DATA STOCKS2; SET STOCKS;
IF STOCK="IBM" AND MONTH(DATE) IN (3,4,5) THEN POST=1;
ELSE IF STOCK="Intel" AND MONTH(DATE) IN (5,6,7) THEN POST=1;
ELSE IF STOCK="Microsoft" AND MONTH(DATE) IN (9,10,11) THEN POST=1;
ELSE POST=0;
IF POST=1 THEN OUTPUT;
RUN;

PROC SQL;
CREATE TABLE STOCKS3 AS 
SELECT STOCK, YEAR, EXP(SUM(LOG(1+RET)))-1 AS RET_POST
FROM STOCKS2
GROUP BY STOCK, YEAR;

SELECT STOCK, MEAN(RET_POST) AS RET_POST
FROM STOCKS3
GROUP BY STOCK;
QUIT;
-----------------------------------------
* SASHELP.STOCKS to find out;
* The total trading volume in each year for each stock;
* The highest price in each year for each stock;
* The lowest price in each year for each stock;
DATA STK; SET SASHELP.STOCKS (KEEP=STOCK DATE LOW HIGH VOLUME);
YEAR=YEAR(DATE);
PROC SORT DATA=STK;  BY STOCK YEAR DATE; RUN;
DATA STK1; SET STK; BY STOCK YEAR DATE;
IF FIRST.YEAR THEN DO;
VOL=0; HIGHEST=HIGH; LOWEST=LOW;
END;
RETAIN VOL HIGHEST LOWEST;
VOL=VOL+VOLUME;
IF HIGH>HIGHEST THEN HIGHEST=HIGH;
IF LOW<LOWEST THEN LOWEST=LOW;
*IF LAST.YEAR THEN OUTPUT;
*KEEP STOCK YEAR VOL HIGHEST LOWEST;
RUN;
PROC PRINT;RUN;

* USE SAS ARRAY AND FUNCTIONS;
DATA PRICE1; SET SASHELP.PRICEDATA;
LOW=PRICE1; HIGH=PRICE1;
ARRAY P(*) PRICE1 - PRICE17;
DO K=1 TO 17;
IF P(K)< LOW THEN DO; LOW=P(K); L=K; END;
IF P(K)>HIGH THEN DO; HIGH=P(K); H=K; END;
END;
RUN;
PROC PRINT DATA=PRICE1 (OBS=20); 
TITLE " HIGHEST AND LOWEST PRICES IN EACH MONTH";
VAR DATE H HIGH L LOW;
RUN;

%MACRO PRC;
DATA PRICE; SET SASHELP.PRICEDATA;
LOW=PRICE1; HIGH=PRICE1;
%DO K=1 %TO 17;
IF PRICE&K<LOW THEN DO; LOW=PRICE&K; L=&K; END;
IF PRICE&K>HIGH THEN DO; HIGH=PRICE&K; H=&K; END;
%END;
PROC PRINT DATA=PRICE (OBS=20); 
TITLE " HIGHEST AND LOWEST PRICES IN EACH MONTH";
VAR DATE H HIGH L LOW;
%MEND;
%PRC; RUN;
-----------------------------------------
* sashelp.pricedata to answer the following questions;
* Create an array for product unit price (Price1 – Price17) ;
* Find the highest and the lowest product unit prices in each month;
DATA PRICE; SET SASHELP.PRICEDATA;
HIGH=PRICE1; LOW=PRICE1;
ARRAY P(*) PRICE1 - PRICE17;
DO K=1 TO HBOUND(P);
IF P(K)> HIGH THEN HIGH=P(K);
IF P(K)< LOW THEN LOW=P(K);
END;
RUN;
PROC PRINT;RUN;
-----------------------------------------
* work.ACT01 using sashelp.pricedata as input;
* array to increase values of price1 theough price17 by 10%;
DATA PRICE; SET SASHELP.PRICEDATA;
ARRAY P(*) PRICE1 - PRICE17;
DO K=1 TO HBOUND(P);
P(K) = P(K)*1.1;
END;
RUN;

* TRY THIS IF YOU WANT TO KEEP THE OLD PRICE VARIABLES;
DATA PRICE; SET SASHELP.PRICEDATA;
ARRAY P(*) PRICE1 - PRICE17;
ARRAY NP(*) NP1 - NP17;
DO K=1 TO HBOUND(P);
NP(K) = P(K)*1.1;
END;
RUN;

/*  %scoreit(ACT01), price2 in obs 5 will be 126.50  */