Thursday 23 August 2012

Bootstrap Methods


The bootstrap is now a vital statistical tool and, due to its power and simplicity, its use is ubiquitous.  Learn more in Dr. Chernick's online course "Bootstrap Methods" at Statistics.com. For more details please visit http://www.statistics.com/bootstrap/.

"Bootstrap Methods" covers the basic theory and application of the bootstrap family of procedures, with the emphasis on applications. After taking this course, participants will be able to use the bootstrap procedure to assess bias and variance, test hypotheses, and produce confidence intervals.  The bootstrap is illustrated also for regression and time series procedures. Basic and improved bootstrap procedures are covered.  The software used is mostly R.

Who Should Take This Course:
Statisticians and data analysts who perform statistical inference, or need to assess uncertainty in their data. Those working with data that does not meet the distributional requirements of standard statistical procedures or with unusual statistics or complex estimators will find the course particularly useful.

Course Program:

Course outline: The course is structured as follows
SESSION 1: Introduction
  • Wide range of application
  • Historical notes
  • Bias estimation
    • Efron's patch data example
    • Estimating other parameters of a distribution

SESSION 2: Parameter Estimation
  • Bias estimation (continued)
    • Error rate estimation problems
  • Confidence intervals and hypothesis test
    • Percentile method confidence intervals
    • Higher order bootstrap confidence intervals
    • A 1-1 relationship between confidence intervals and hypothesis tests
    • Problems with bootstrap confidence intervals for variances

SESSION 3: Regression, Time Series, Which Methods?
  • Linear Regression, bootstrap residuals or vectors
    • Non-linear Regression
      • A Quasi-optical experiment
    • Nonparametric Regression
      • Cox Model
      • CART
      • Bootstrap Bagging
    • Time Series Analysis
      • Model-based vs block resampling
    • Bootstrap variants
      • Bayesian bootstrap
      • Smoothed bootstrap
      • Parametric bootstrap
      • Iterated bootstrap
    • Number of repetitions (replications)

SESSION 4: Special Topics, Bootstrap Failures and Remedies
  • Spatial data: kriging
  • Subset selection
    • Examples of Gong and Gunter
  • p-value adjustment
  • Process capability indices
  • Bioequivalence
  • Failure Due to Small Sample Size
  • Failure Due to Infinite Moments and Remedy (introducing m-out-of-n bootstrap)
  • Failure Due to Estimating Extremes and Remedies

Dr. Michael Chernick and Dr. Robert LaBudde are the coauthors of "Bootstrap Methods with Applications in R," the course text.  Chernick studied with Bradley Efron in the bootstrap's early days, is a Fellow of the American Statistical Association, the 1983 winner of the Wolfowitz Prize, and the author of more than 30 journal articles.  LaBudde is president and founder of Least Cost Formulations, Ltd., a mathematical software development company specializing in optimization and process control software for manufacturing companies.

You will be able to ask questions and exchange comments with Dr. Michael Chernick and Dr. Robert LaBudde via a private discussion board throughout the course.   The courses take place online at statistics.com in a series of 4 weekly lessons and assignments, and require about 15 hours/week.  Participate at your own convenience; there are no set times when you must be online. You have the flexibility to work a bit every day, if that is your preference, or concentrate your work in just a couple of days.

For Indian participants statistics.com accepts registration for its courses at special prices in Indian Rupees through its partner, the Center for eLearning and Training (C-eLT), Pune.

For India Registration and pricing, please visit us at www.india.statistics.com.

Call: 020 66009116

Websites:

No comments: