Ghement Statistical Consulting Company Ltd.
  Home > Training > Take Your R Skills to the Next Level - August 4, 2014 - Boston, USA
Ghement Statistical Consulting Company Ltd.
301-7031 Blundell Road
Richmond, British Columbia
Canada, V6Y 1J5
Tel: 604-767-1250
Fax: 604-270-3922
E-Mail: info@ghement.ca


Isabella R. Ghement 2016


































































































































































































 

Temporal Trend Detection and Analysis in the Environmental Sciences Using the Open-Source Statistical Software R


We are pleased to announce the 3-day course "Temporal Trend Detection and Analysis in the Environmental Sciences Using the Open-Source Statistical Software R" on May 13, 14 and 15, 2015 in Vancouver, B.C., Canada. This course is ideal for participants who are interested in expanding their toolkit for investigating and quantifying temporal trends present in data collected over time as well as those who are interested in producing forecasts based on such data.

If you are interested in attending this course, would like to receive further details and be added to the list of participants, please e-mail the course instructor, Dr. Isabella Ghement, at isabella@ghement.ca or call her at 604-767-1250 prior to the registration deadline of May 8, 2015.

Alternatively, if you would like us to offer this course privately at your premises or provide you with information on our other course offerings, please let us know.


Course Description


Temporal trend detection and analysis is an important endeavour in the environmental sciences, as it facilitates monitoring and management of natural resources. In this 3-day course, you will learn a variety of statistical methods for detecting and analyzing temporal trends and become familiar with their real-world applicability using the open-source statistical software R.

On Day 1, you will learn regression methods for deterministic trend testing and estimation, including linear regression modeling with/without serially correlated errors, generalized linear regression modeling (e.g., Poisson, Negative Binomial) and generalized additive modeling. These methods are useful for capturing parametric/nonparametric deterministic trends in the values of an outcome variable, while also accounting for the potential influence of relevant factors on these trends.

On Day 2, you will learn distribution-free methods for testing monotonic trends such as the Mann-Kendall test, the seasonal Kendall test and the regional Kendall test, which are handy when the data do not follow standard assumptions (e.g., normality). You will also learn additional regression modeling methods for trend detection and analysis such as Theil-Sen regression, quantile regression and change-point regression.

On Day 3, you will learn time series modeling tools for uncovering stochastic trends and exploiting them for forecasting purposes: autoregressive and moving average (ARMA) models, integrated ARMA models (ARIMA), seasonal ARIMA models and exponential smoothing models. In particular, the entire cycle of the forecasting process will be emphasized: forecasting model formulation, forecasting model verification and refinement, point forecast production, interval forecast production via time series bootstrapping, retrospective point forecast evaluation, forecasting diagnostics, as well as presentation and interpretation of forecasting results.

By the end of the course, you will have a working knowledge of a series of modern and valuable statistical approaches for trend detection and analysis and be able to apply them to your own data using the open-source statistical software R and the R scripts provided in the course.


Benefits to Participants

After attending this course, participants will be able to:

  • Formulate, estimate and validate appopriate regression models for uncovering temporal trends present in the data;
  • Accommodate special features of the data (e.g., seasonality, autocorrelation, non-detects) when estimating trends via regression models;
  • Formulate, estimate and validate time series models for capturing stochastic temporal trends present in the data and use these models as a basis for forecasting.


Course Outline


Day 1

  • Linear Regression with Uncorrelated Errors
  • Linear Regression with Correlated Errors
  • Generalized Linear Regression
  • Generalized Additive Modeling

Day 2

  • Mann-Kendall Test and Theil-Senn Regression
  • Seasonal Kendall Test
  • Regional Kendall Test
  • Quantile Regression
  • Change-Point Regression

Day 3

  • ARMA Models
  • ARIMA Models
  • SARIMA Models
  • Exponential Smoothing Models
  • Forecasting


Course Format


This course is limited to 18 participants and consists of a series of short lectures and demonstrations followed by hands-on, interactive sessions for the participants. Each participant will be provided with:

  • A bound copy of the Course Notes;
  • A CD-ROM containing all examples and exercises used during the course;
  • 30 days of free course-related technical support following the course.

Course Leader


The course is led by Dr. Isabella Ghement. Isabella obtained her Ph.D. in Statistics from the University of British Columbia (UBC) in 2005. Isabella has presented numerous public and private workshops/courses on the statistical software package R to researchers, graduate students, government agencies and corporations in Canada. She also lectured part-time on advanced statistics at the Sauder School of Business, UBC between 2005 and 2012. Isabella is actively engaged in statistical consulting through her company Ghement Statistical Consulting Company Ltd. Her statistical consulting clients include federal and provincial government agencies, contract research organizations and academic researchers. Isabella co-authored the following publications based on her Ph.D. work on nonparametric regression: "Seasonal Confounding and Residual Correlation in Analyses of Health Effects of Air Pollution" (Environmetrics. 2007; 18(4): 375-394) and "Robust estimation of error scale in nonparametric regression models" (Journal of Statistical Planning and Inference. 2008; 138(10): 3200-3216). Most recently, Isabella has acquired statistical expertise in the field of mixed treatment comparisons, a generalization of meta-analytic methods allowing for the comparison of multiple medical interventions with respect to their efficacy and safety. In this field, Isabella has co-authored publications such as "Incorporating multiple interventions in meta-analysis: an evaluation of the mixed treatment comparison with the adjusted indirect comparison" (Trials 2009, 10:86 doi:10.1186/1745-6215-10-86), "Estimating the Power of Indirect Comparisons: A Simulation Study" (PLoS ONE 6(1): e16237. doi:10.1371/journal.pone.0016237) and "Multiple treatment comparison meta-analyses: a step forward into complexity" (Clinical Epidemiology. 2011; 3: 193-202).


Prerequisites


Participants should have some basic knowledge of R and statistics.

Participants should bring a laptop computer pre-installed with the open-source R software and its user-friendly interface R Studio.

R can be installed from http://cran.stat.sfu.ca and R Studio can be installed from http://www.rstudio.org.

In addition to R and R Studio, participants will be required to install several R packages on their laptops. Instructions on R package installation will be distributed to participants approximately one week prior to the course.


Registration Form


The course registration form is available for download via the following links:

To register for this course, please complete and sign the above registration form and send it to us by e-mail at info@ghement.ca or fax it to us at 604-270-3922 at your earliest convenience.


Location


Downtown Vancouver in the new British Columbia Institute of Technology (BCIT) Building, Room 410, 555 Seymour Street, Vancouver, B.C.
(Course not affiliated with BCIT.)

Instructions for getting to the course venue are available here: http://www.ghement.ca/RCourseMay2015.html

You can also use the links below to plan your visit to the BCIT Downtown Campus:


View larger map


Dates and Times


Dates: Wednesday, Thursday and Friday, May 13, 14 and 15, 2015

Registration: On all days, course registration and set-up begin at 8:30am.

Time: The course starts at 9:00am and ends at 5:00pm on each day.


Cost


The attendance fee for the course is $885.00 plus 5% GST per participant and includes a bound copy of the Course Notes, a CD-ROM containing all course examples and exercises and 30 days of free course-related technical support following the course. The attendance fee also includes morning and afternoon coffee, tea and snacks.


Discounts


Groups of 3 or more participants from the same organization receive a 10% discount.


Cancellation Policy


  • 100% refund if written notification of cancellation is received by May 1, 2015. Please note that no refunds will be issued after this date.
  • In the event you become unable to attend after the May 1, 2015 refund deadline, you may delegate a substitute attendee. Please notify us of any changes as soon as possible via e-mail at info@ghement.ca or telephone at 604-767-1250 or fax at 604-270-3922.

  • The course is limited to 18 participants per day so we encourage you to register early. The registration deadline is May 8, 2015.
  • To reserve your place, please follow the instructions below:

    1) Pre-register by e-mailing us at info@ghement.ca, or telephone us at 604-767-1250, or fax us at 604-270-3922.

    2) Complete the Registration Form.

    3) E-mail or fax the completed Registration Form to us, and mail your cheque payable to Ghement Statistical Consulting Company Ltd.; or request us to invoice your organization where indicated on the Registration Form.

  • Your reservation will be confirmed via e-mail as soon as we receive your registration form.
  • Please do not make any travel arrangements until your reservation has been confirmed by us in writing via e-mail.

 

Home | About Us | Consulting | Training | Case Studies | Testimonials | Resources | Contact Us | Privacy Policy