1. K-State home
  2. »Arts and Sciences
  3. »Statistics
  4. »News and Events
  5. »Workshops - Summer 2018

Department of Statistics

Workshops - Summer 2018

The Department of Statistics will host four distinct workshops that focus on recent developments in statistical methodology and computational techniques.  These workshops are targeted for K-State faculty and affiliated researchers, including collaborators, graduate students and postdocs. Participants should have sufficient statistical background; specific requirements can be found in the description of each workshop.

All workshops will be conducted during Summer 2018 at the K-State Manhattan campus.  Attendance is free for KSU affiliates, but pre-registration is required. Space is limited, and qualified registrants are accepted on a first-come first-served basis.  
Shell Oil logo  

These workshops are supported by Shell Oil.     




Introduction to Spatial Analysis in R

photo of earth from space

Department of Statistics Workshop Co-sponsored by Shell Oil 

Date: May 18, 2018, 8 am to 6 pm
Location: Kansas State University, Dickens Hall, Room 207
Presented by:

  • Dr. Trevor Hefley, KSU Statistics
  • Dr. Mevin Hooten, Colorado State University & USGS
  • Haoyu Zhang, KSU Statistics Graduate Student
  • Nelson Walker, KSU Statistics Graduate Student

No prior experience with R is required, but you will need to bring a laptop that has R installed.  A list of required R packages will be emailed to the participants before the workshop.

Capacity: Limited to 24 participants.

Download the PDF for more information.

This workshop is at capacity, and no new applications are being accepted. The full course material will be offered in STAT 764 during the fall semester.

Learn more about STAT 764

Course description for STAT 764:  Applied Spatio-Temporal Statistics.

Construction and analysis of spatial, time-series, and spatio-temporal data sets. Topics includes data generation using geographic information systems, exploratory data analysis and visualization, and descriptive and dynamic spatio-temporal statistical models. For context, a focus will be on biological or ecological data.  (3 credit hours) 
Prerequisites: STAT 510 or 770, STAT 705 or 713, and STAT 726 or equivalent

View the course schedule for Fall

 


Statistical Machine Learning for High Dimensional Data

heat mapDepartment of Statistics Workshop Co-sponsored by Shell Oil 

Date: June 1, 2018, 8 am to 5 pm
Location: Kansas State University, Dickens Hall, Room 207
Presented by:

  • Dr. Cen Wu, KSU Statistics
  • Guotao Chu, KSU Statistics Graduate Student
  • Fei Zhou, KSU Statistics Graduate Student

This workshop is targeted for K-State faculty and affiliated researchers, including collaborators, graduate students and postdocs with an interest in statistical machine learning for high dimensional data.  No prior experience programming with R is required. However, some familiarity with R will be very helpful. You will need to bring a laptop with power cable. Please have R software installed on your laptop before the workshop. We will use R Studio for illustration. No prior biology experience is required. We will focus on high dimensional cancer genomics data but the methodology can be readily extended to other domains. A list of required R packages will be emailed to all participants at least one day before the workshop.

Capacity: Limited to 20 participants. 
Please contact Dr. Cen Wu to RSVP (wucen@ksu.edu).

Download the PDF for more information.

This workshop is at capacity, and no new applications are being accepted.  The full course material will be offered in STAT 766 in Fall 2019.

Learn more about STAT 766

Course description for STAT 766: Applied Data Mining/Machine Learning and Predictive Analytics.

This course addresses the complete process of building analytical tools suitable for learning from data, including automatic online data collection, feature extraction, supervised and unsupervised statistical machine learning methods, evaluation, and report writing. Automatic retrieval of various format online data, including JSON, REST and Streaming API, http(s), html, xml, and databases. Statistical text processing/mining, state of the art top performers of supervised and unsupervised data mining methods,  case studies and applications to business, government, social and news media data. Methods include regularized linear and logistic regression, classification trees, nearest neighbor methods, support vector machines, Naive Bayes, random forests, boosting/bagging/AdaBoost, clustering, latent Dirichlet allocation, network analysis, and topic modeling models.

The course is 3 credit hours and is offered in the Fall semester of odd years.
The pre-requisites are STAT 705 or 713 or 717, and prior computer programming proficiency on C, C++, Fortran, R or Python (e.g., CIS 209, STAT 726)



Applied Classical and Modern Multivariate Statistical Analysis

hypercubeDepartment of Statistics Workshop Co-sponsored by Shell Oil 

Date: August 15, 2018, 8 am to 6 pm
Location: Kansas State University, Dickens Hall, Room 207
Presented by:

This workshop provides a relatively broad introduction to commonly used techniques for analyzing multivariate data. Basic exploratory tools for summarizing and visualizing multivariate data will be introduced, and the classical multivariate statistical inference procedures will be covered, such as multivariate mean comparison, tests on covariance matrix from multivariate normal distribution, correspondence analysis, principal component analysis and classification. If time permits, this workshop will also discuss some modern multivariate analysis methods for regression and classification, such as support vector machines and tree-based methods. Statistical theories for all topics will be kept at a minimum level, and extensive examples will be used to illustrate how to implement the relevant statistical procedures using R.

Capacity: Limited to 30 participants.  
Please contact Dr. Weixing Song to RSVP (weixing@ksu.edu)

Download the PDF for more information


Mixed Models for Agricultural and Biological Research

cows at a feed lot

Department of Statistics Workshop Co-sponsored by Shell Oil 

Date: Early August 2018, exact date and location TBD
Instructor: Dr. Nora M. Bello, KSU Statistics 

This workshop will provide a fairly comprehensive exposition of mixed-model based statistical data analysis, power determination and sample size calculation for commonly used experimental designs in the agricultural and biological sciences. The approach will be workshop-like, example-driven and primarily based on the various mixed model analysis procedures available in the SAS software. Specification and implementation of general and generalized linear mixed models for normal and non-normal responses will be discussed in the context of hierarchically structured data. Specific topics will include complete and incomplete blocks, nested effects, subsampling and repeated measures, as well as diagnostics for mixed models.

(Registration is not yet open for this workshop.)

Please note: Dr. Bello is on sabbatical through Spring 2018.  She respectfully requests that you postpone contacting her about this workshop until she returns in mid-summer.
As more information about this workshop becomes available, it will be posted on the K-State Statistics Department website (www.k-state.edu/stats/news).