Homework 1 
 Review of Dates, cubic splines, merging 

 Questions 

(note: Click on SAS Code at the bottom to see the initial 
SAS code for this problem - copy and paste it into SAS then 
modify as needed)

 
  In the SAS Code link below there are some (artifical) data 
on calls to a shipping firm.  One dataset has calls, and the 
next has shipments. The dates are not exactly the same as you 
can see. 

Run the program unmodified.  From the correlations, 
and keeping in mind the offset, what do you think is a
typical delay (in days) between an order being called
in and the resutling shipment? Plot the graph of shipments
and versus the lagged calls (lagged as you decided above)
Notes:
 Y=lag(X) sets Y to the X from the previous record
    
 Y=lag3(X) sets Y to the X 3 records back. 
    
 @@ leaves the pointer where it is until line is 
            exhausted (optional: try removing it and see what 
            happens)
    
 See class notes or suggestions below for plot options.

 Using the information in the class notes, create date 
variables for both data sets.  Then merge by date as suggested
(remove the * from the by date statement) Plot calls and 
shipments against date. 

   Note: 

  You could use (for generic Y and X variables) 
          PROC PLOT; PLOT X*DATE="*" Y*DATE="+"/OVERLAY;    
                     or
      PROC GPLOT; PLT (X Y)*DATE/OVERLAY; 
			  SYMBOL1 V=NONE I=JOIN C=GREEN;
                    SYMBOL2 V=NONE I=JOIN C=BLACK; 


 Now, either before or after merging, use PROC EXPAND to fill 
in some values for shipments where they are missing. In this way, 
get a dataset with date, calls and shipments where missing shipments 
have been imputed. Repeat the correlation and the plot from part (1) 
with the newly completed shipments data.  Comment on the effect of 
the imputations. 

 Give the equation that is used to interpolate between the 5th 
and 7th value of shipments.  

 Explain what these are doing: 
  (obs=10) 
    
  the label option in proc print
    
  call1-call3  (versus call1 call3 without the "-") 



 (optional) 
  Try this in the merged dataset
    PROC ARIMA; I VAR=SHPMNTS CROSSCOR=(CALLS); RUN;


  SAS Code
   output