Welcome to Stat 740, Statistical Computing. This page contains updates to the course syllabus, class notes with annotation and homework assignments. I hope to have both HTML and postscript versions of the class notes available. Send any questions to grego@stat.sc.edu

- Notices
- The final exam is due 5 PM Monday May 5.

- The final exam is due 5 PM Monday May 5.
- Syllabus
- Homework
- Web resources
- Brian Habing's Spring 2004 STAT 740 course
- Jim Gentle's Statistical Computing course.
- Mervyn Mersinghe's R course
- Nancy Reid's R statistical programming/applications course

- Brian Habing's Spring 2004 STAT 740 course
- Handouts and Exercises
- Initial notes on "sourcing" a file
- Instructions on calling a Fortran program in Splus on a Unix machine
- Brian Habing's Instructions on calling a Fortran program in R on a PC. These are also available from his web page
- Additional instructions on calling a Fortran program in R on a PC
- Notes on kernel density estimate commands from class.
- Computer exercise to estimate the parameters in a zero-inflated Poisson random sample.
- Handout showing steps in computing the gradient, hessian and information matrix for a three-parameter gamma distribution.
- Handout showing the EM step increases the observed data log likelihood.
- Handout with score vectors and information matrices for EM example.
- Computer exercise to explore R bootstrap commands.
- Computer exercise for MCMC in R.

- Instructions on calling a Fortran program in Splus on a Unix machine
- R code and notes
- Samples of student code
- Buffon's needle simulation
- Hit or Miss graphical demo from STAT 517.
- Program to estimate a Cauchy tail probablity using Crude Monte Carlo integration.
- Importance sampling simulation
- Plotting commands for importance sampling simulation
- Box-Muller commands from class.
- Program to sample standard normal random deviates using acceptance sampling from an exponential distribution.
- Program to sample truncated standard normal random deviates using acceptance sampling from a truncated Weibull distribution.
- R code to produce a least squares cross validation plot for kernel density estimation.
- R code to compute machine epsilon for use in optimization algorithms.
- R code to demonstrate use of uniroot to find a mixed normal quantile.
- R code to estimate the parameters in a zero-inflated Poisson random sample.
- SAS NLMIXED code to estimate the parameters in a zero-inflated Poisson random sample.
- R code to estimate the parameters for a mixed normal distribution using the EM algorithm and generate a gradient trace contour plot.
- R code to estimate the mean of exponential data with left- and right-censoring.
- R commands and bootstrap function for blood serum example. Includes a second set of commands to be used for the bootstrap confidence interval exercise.
- R commands and bootstrap function for bootstrap correlation example. Includes a second set of commands to be used for the bootstrap confidence interval exercise.
- R commands and bootstrap function for bootstrap loess example.
- Simple Gibbs Sampler demo for one-way random effects model.
- R commands and data set for simple Gibbs sampler class demonstration.
- R commands for in-class demo of a multi-chain Gibbs sampler for the one-way random effects model.
- Gibbs sampler and Metropolis-Hastings algorithms for the Ising model.
- In-class demo of non-converging Gibbs Sampler.
- Model file , data file , and initial values file for one-way random effects BRugs demo.
- Generating bivariate normal rvs using a Metropolis Hastings algorithm.
- ARS algorithm code with in-class demo of poor choice of grid. (A successful demo should soon be posted)

- Samples of student code
- Midterm Exams
- Project Guide
- Report guidelines
- Modification of John Spurrier's Oral Report Guidelines
- An Alternative Set of Oral Report Guidelines

- Report guidelines
- Final Exams
- 2006 Final Exam
- 2008 Final Exam with script for Q2.

- 2006 Final Exam

Return to Department of Statistics Home Page