Specifications:

You are allowed at most three DATA steps and eight PROC steps to complete this assignment.

Create library and file references that point to the provided data sets on your class' shared drive. As discussed in

class, name the libref InputDS and the fileref RawData. Additionally, create a library named HW$2$ that points to the

location where you will store any data sets you create. Save any results you produce to the same directory as

your HW$2$ data sets.

No changes from how we set these up in HW$1$

Recall, we do not have access to your $S$ drive. All you do is submit the code and we run it using the paths we

want to use on our S drive.

We will be using the BasicSales data this week. The data is spread across three plain text $($i$.$e$.,$ raw$)$ files.

BasicSalesNorth contains all the records with Region $=$ North

BasicSalesSouth contains all the records with Region $=$ South

BasicSalesEastWest contains all the records with Region $=$ East OR Region $=$ West

The file contents are similar $-$ each begins with a header that provides specific information on the file before

presenting the data $-$ but the structure is slightly different in each file.

Read in each raw file into a data set with a corresponding name. E$.$g$.,$ BasicSalesNorth.dat should be placed into a

SAS data set named BasicSalesNorth.

My data sets are provided in the usual place on the shared drive

You should match my data sets exactly. That means metadata and data. For now you are still going to have to

compare these manually, but the data sets are still small so this should relatively be straightforward. Pay attention

to both the variable attributes $($name$,$ order, type, format, length, etc.$)$ and variable values.

After reading the data in$,$ produce two reports that match mine: HW$2$ Duggins Basic Sales Report.pdf and HW$2$

Duggins Basic Sales Metadata.rtf$.$ Name your files as I did, replacing your last name for mine. As will be standard for

us$,$ I've listed a few items below that you cannot easily tell from the reports themselves.

The PDF uses the Journal style and the RTF uses the Sapphire style.

Titles use a $14$ pt font, subtitles use $10$ pt font, and footnotes use $8$ pt font.

Pay close attention to the PDF $-$ the journal style depends on subtle changes $($like italics$)$ to convey information

in the output objects. $($This is because academic journals don't typically allow for color printing or anything else

that looks even vaguely interesting.$)$

The two reports are not identical. The RTF includes some results that are excluded from the PDF and vice versa;

the RTF is limited to results specifically about the North and South regions. You already know how to identify

and select$/$exclude output objects $-$ but here you need to selectively place them in a particular destination. To

do this, it might be helpful to use the following statements in some cases. These work with any external output

destination we use $($WORD$,$ POWERPOINT, etc.$)$ but I've shown them here for the PDF destination.

ODS PDF EXCLUDE NONE;

ODS PDF EXCLUDE ALL;

In the MEANS output I have applied a custom format to the TotalDue variable. That format is named BasicAmt$-$

Due and is located in the InputDS library. As discussed in class and in your reading, it is possible to use a format

from a permanent library. To use it in this assignment, you'll need to do the following:

Once you establish the InputDS library as usual, you need to tell SAS you want to access the formats

stored there. To do this, include an OPTIONS statement that includes FMTSEARCH $=($InputDS$)$ after you've

established this library.

When printing out formats, you can choose which formats are included in your printout by using the SELECT

statement and naming the format$($s$)$ you want to see.

Do not redefine the format yourself! The goal here is to use one that has been provided to you! PROGRAMMING HW #$2$

The labels are rather long, and make the PROC MEANS results look pretty messy. Just like PROC PRINT has a LABEL

option, PROC MEANS has a NOLABELS option $-$ use it$!$

In the PROC FREQ step, the dates have been grouped into quarters labelled with Roman numerals using the QTRR

format. $[$Optional task: Compare the default sort order to the results when using ORDER$=$FORMATTED. Why does

this happen?$]$