FALL QUARTER, 2014
October 3, 12:00-1:30pm
Users typically need to read input data, shape the data, apply an analysis routine, and then output the results.
“Leaping the Hurdles and Navigating the Maze: Getting NIH Funding”
October 23, 11:00 AM – 12:00 PM
This is an introduction to the culture of the US National Institutes of Health. I will talk about how the NIH functions, describe the process of an award from application to review to funding and provide some suggestions on grant writing including mechanisms, tips and things to avoid. There will be time for questions.
“How to Effectively Talk About Your Research with Diverse Audiences”
December 10, 11:00 AM – 12:00 PM
When you are asked to talk about your research, many of the challenges are the same no matter who your audience and what your communications medium (PowerPoint or not). How to get and keep the audience on board. How to choose what to say and what to leave out.
How to identify and reinforce take-away messages.
With your peers, you can assume some shared understanding and common language. Start at Point C and with your help the audience will hop on board and follow along. But what are the pitfalls?
How can your presentation be more effective?
Non-academic audiences pose other challenges. How to establish context and explain complicated concepts. How to tweak main messages for high-stakes audiences such as policymakers and the media.
WINTER QUARTER, 2015
“Identifying and Accessing Datasets for Studies on Health and Aging”
March 13, 12:00 AM – 1:00 PM
This presentation outlines the general approach to identifying and accessing datasets for secondary data analyses related to health and aging. Within this framework, we will outline the services provided by the UCLA Older American Independence Center’s Data Access Pilot Project (DAPP). We will provide a brief overview of several datasets, including MIDUS (Mid-Life in the US), MESA (Multi-ethnic Study of Atherosclerosis) and SWAN (Study of Women’s Health across the Nation).
SPRING QUARTER, 2015
An increasing number of longitudinal datasets are being made available. The longitudinal nature of the dataset may be represented as a hierarchy of stages, say, measurements across time nested within an individual. We’ll discuss how hierarchical models account for the nested structures and how Generalized Linear Models account for different outcome-response data types. Through hands-on exercises, the workshop will give a brief overview of the motivation and intuition of longitudinal data analysis.
Stan is an open-source, Bayesian inference tool with interfaces in R, Python, Matlab, Julia, Stata, and the command line. Users write statistical models in a high-level statistical language. The default Bayesian inference algorithm is the no-U-turn sampler (NUTS), an auto-tuned version of Hamiltonian Monte Carlo. Stan was developed to address the speed and scalability issues of existing Bayesian inference tools. The goal of the workshop is the practical application of Stan to different models starting with ordinary linear regression and ending with more complex models such as generalized linear mixed and hierarchical models.