Course (Section)

STA 402/502 (A)

Meeting Time:

800-850 MWF (plus other make-up times to be arranged in consultation with students)

Meeting Location

106 Bachelor Hall

Prerequisites:

STA 401/501; STA 671; or permission of the instructor. Willingness to work.

Professor:

Dr. John Bailer

E-mail:

baileraj@muohio.edu

URL:

http://www.users.muohio.edu/baileraj

Office (phone)

292 or 122B Bachelor Hall (529-3538)

FAX: 529-1493

Office Hours

10:00 - 11:30 M W F

(other hours by appointmentdon't be shy!) 

MU Calendar

# weeks

Tentative topics (associated with estimated # weeks!)

1

BASIC CONCEPTS (Ch. 1)

1-2

Constructing a data set for analysis:  reading, combining, managing and manipulating data sets (Ch. 2)

1.    Temporary versus permanent status of data sets – LIBNAME
2.    Reading data into a SAS data set
2.1  Reading data directly as part of a program – anyone for datalines?
2.2  Reading data sets that were saved as text – infile can be your friend
2.3  Sometimes variables are in particular columns or in particular formats
2.4  Reading comma-separated values – text files with “,” delimiters
2.5  Reading Excel spreadsheets directly
2.6  Reading SPSS data files – the little (SPSS) engine that could
3.     Writing out a file or constructing a simple report
4.     Concatenating data sets/adding observations – SET
5.     Merging data sets/adding variables – MERGE
6      Database processing with PROC SQL

1-2

Using SAS procedures (Ch. 3)

1    SAS system options – options
2    Statements that can modify output of most procedures –TITLE, LABEL, FORMAT
3    Defining your own formats for variable values
4    Selecting or stratifying an analysis by values of a variable – WHERE, BY, SORT
5    Displaying data set contents and properties – PRINT and CONTENT
6    PROC PRINT for listing the observations in a data set
7    Basic graphical displays
8    Using Scatter plots to display relationships between numeric variables
9    Summarizing categorical variables - FREQ
10  Summarizing numeric variables – UNIVARIATE, MEANS
11  Selecting a simple random sample - SURVEYSELECT
12  Randomly assigning treatments to observations - PLAN

1

Complex table construction and output control (i.e., “pretty” output) (Ch. 4)

1.  PROC TABULATE
2. Building from simple specifications:  nitrofen data
3. Enhancing PROC TABULATE output
4.  Output Delivery System (ODS)
4.1 Basic Ideas
4.2 Favorite Destinations – RTF, HTML and even PDF
4.3 What’s produced (i.e. output objects) and how to select them
4.4 Another Destination that Stat Programmers should visit – OUTPUT

1

Basic models in SAS (Ch. 5)

1.  Overview of modeling
2.   Linear Regression models – REG, GLM
2.1  Example 1:  Motorboats and manatees – a look at simple linear regression
2.2 Example 2:  Big brains and big bodies – specifying and fitting a multiple regression model
3.     ANOVA models – GLM for one-way anova
3.1 Example 3:  Rotting meat – package comparisons tested with one-way anova model
4.     ANOVA models – GLM for anova model with two or more factors

1-2

Producing Statistical Graphics (Ch. 6)

1.  Old School (device-based) / New School (template-based) SAS graphics
2.   ODS stat graphics (New School)
3.   Modifying graphics using the statistical graphics editor (New School)
4.   Graphing with style (and templates) (New School)
5.   Statistical Graphics – entering the land of SG* (New School)
                  5.1  Case Study using SG* graphics
6.   Back to (old) school (graphics)
7.   Customizing graphics (Old School)
8.   Why you need to learn about annotate data sets (Old School)
9.   Case study:  comparing distributions of responses (Old School)
10. Descriptive displays of spatial data (Old School)

1-2

Formatting, basic DATA step manipulations and programming (Ch. 7)

1.  Internal representations and output displays
2.  Character, numeric and date formats
3.  Recoding and transforming variables in a DATA step
4.  Ordering how tasks are done – precedence of operations
5.  What goes and what stays in a data set – DROP, KEEP, IF, WHERE, OUTPUT
6. Structured thinking about writing programs – pseudo-code and modules
7. CASE STUDY 6.1:  Is the two-sample t-test robust to heterogeneous variances?
8. CASE STUDY 6.2:  Monte Carlo integration to estimate Pr(0<Z<1.645) for Z~N(0,1)
9. CASE STUDY 6.3:  Simple percentile-based bootstrap
10.Throw out your tables of statistical distributions – CDF, PDF, QUANTILE
11. Generating variables using random number generators – RAND

1-2

Programming in a DATA step (Ch. 8)

1. Storage bins for collections of values - ARRAYS
1.1 Example 1: Defining values in ARRAY variable list directly
1.2 Example 2: Inputting values in ARRAY variable list
1.3 Example 3: Changing missing value codes for numeric variables to “.”
1.4 Example 4: Recoding missing values for all numeric and character variables
1.5 Example 5: Creating multiple observations from a single record
2. Case Study 1: Monte Carlo P-value for test of spatial randomness
3. Remembering variable values across observations – RETAIN
3.1 Example 6: Processing multiple observations for an individual
4 Case Study 2: Randomization test for the equality of two populations

1

MACRO programming (Ch. 9)

0.   What is a macro and why would you use it?
1.   Motivation for Macros:  numerical integration to determine P(0<Z<1.645)
2.   Macro processing
3.   Macro variables
4.   Conditional execution, looping and macro programs
5.   Debugging macro coding and programming
6.   Saving macros - %include +autocall+stored compiled macros
7.   Functions/routines of potential interest to macro programmers - %index, %length, %eval, symput, symget

1-2

Programming with matrices and vectors – IML (Ch. 10)

2-5

TOPICS IN STATISTICAL PROGRAMMING (varies)

* Introduction to quantitative programming in R (objects-vectors, lists, matrices, dataframes; reading data [scan, read.table, sas.get]; summarizing data sets [mean, var, summary, table]; graphical displays [plot, pairs, coplots]; writing functions.
* Intro., packages & GUI (Rcmdr)
* Data structures
* Basic graphics
* Programming (flow control, functions, etc.)
* Simulation
* Other topics?