A practical and example filled tour of simple and multiple regression techniques (linear, logistic, and Cox PH) for estimation, adjustment and prediction.

Loading...

来自 约翰霍普金斯大学 的课程

Statistical Reasoning for Public Health 2: Regression Methods

72 评分

A practical and example filled tour of simple and multiple regression techniques (linear, logistic, and Cox PH) for estimation, adjustment and prediction.

从本节课中

Introduction and Module 1A: Simple Regression Methods

In this module, a unified structure for simple regression models will be presented, followed by detailed treatises and examples of both simple linear and logistic models.

- John McGready, PhD, MSAssociate Scientist, Biostatistics

Bloomberg School of Public Health

So greetings and welcome to Statistical Reasoning Two.

I just wanted to take a moment to briefly and warmly welcome you

back after your well deserved break at the end of Statistical Reasoning One.

Well anyway, despite the fact there was no break we'll just capitalize on

the momentum that you had coming out of Statistical Reasoning One.

And in this term we'll expend the ideas and

methods we've worked so hard to develop in Statistical Reasoning One,

and we're going to be doing things under an umbrella cache of methods called

regression methods, which will allow us to do everything that we've showed how to

do in Statistical Reasoning One, but take them one or more steps further.

So we'll be looking at these regression methods which will allow us to

model outcomes, the kind we were looking at in Statistical Reasoning One,

continuous, binary and time to event.

As a function of a single predictor,

which we've already looked at in Statistical Reasoning One,

with things like the t test, or the chi squared test, and

the resulting estimated mean differences, risk differences, relative risk, etc.,

and we also did similar things with time and event data.

But what we'll be expending these methods to do is allow us to look at

these outcomes as they associate with more than one predictor at a time.

So not only will we be able to better predict the functions of our

outcomes by using more information potentially in data sets we get, but

this will also give us an excellent and straightforward method for

do things like adjusting associations and comparisons of interest in the presence of

potential confounders and estimating easily, relatively easily,

different associations between the outcome and predictor for different sub groups of

our data, all while adjusting for other factors, that may be related

these associations, via their associations with the predictor and the outcome.

So anyway I'm thrilled to have you back in the class.

I'm looking further, forward to the journey we'll take this term.

So onward and upward as we move forward with Stats Reasoning Two.