Материал готовится,

пожалуйста, возвращайтесь позднее

пожалуйста, возвращайтесь позднее

[NOISE] Okay, welcome everyone to this

course on Applied Logistic Regression.

This is an eight week long course.

And we are going to be going

through Approximately four of

the chapters in the textbook that

the course is based on and named after,

called Applied Logistic Regression.

Where I am one of the authors,

along with my colleagues,

David Hosmer and Rodney Sturdivant.

For those of you who are interested

in the book, you may find

the book is available, I think,

anywhere you can get books.

Again, the title is

Applied Logistic Regression.

The third edition was released in

March of 2014, March of 2013.

So, it's relatively new, and again,

we will be covering perhaps four or

so chapters of this book.

If you get interested

in logistic regression,

I strongly recommend that at some point

you get the book and you can find

many more topics beyond what we'll be

able to cover together in this course.

So we we are going to do a lot together.

Certainly, the lectures are going to take,

are going to take us time to go through.

But you really will learn this course,

you'll really learn to apply

logistic regression models

by doing the homeworks that you'll

find at the end of each of the weeks.

And I've given you, and

you'll find on the website,

you'll find the the,

the homework assignments.

And you will also find solutions

to the homework assignments.

And, it's, the expectation is that you

will download the statistical package

STATA, into your computers, and that you

can carry out the assignments using STATA.

And it will make it easier if you

use STATA because I've given you all

the solutions as well.

So the idea here is not to have

you become a STATA expert,

but the idea is to give

you practical experience,

running logistic regression models on,

with the computer program.

And interpreting the output

of those programs.

And you can do that very easily

with STATA with this course.

But if you prefer to use SPSS,

or SASS, or R, or

anything else that you're used to,

you can certainly do that.

The material in this course is such that

most of what we do is available in

all modern statistical packages.

So you have the option to do

that yourself if you would like.

So, we'll start right away with

logistic regression analysis, and

the goal of a logistic regression

analysis is to find the best-fitting,

simplest model that we possibly can.

That describes the relationship between

an outcome or dependent variable or

response variable and a set of independent

variables which we call predictors or

explanatory variables.

We also call them covariates.

But what distinguishes a logistic

regression model from a linear regression

model, is that the outcome variable

in linear regression is binary or

dichotomous.

Now actually, if you did get our book and

you looked at some of the later chapters

of the book, you realize that,

that's not even a 100% true statement.

Because in the more advanced

logistic regression modeling,

you may have an outcome variable that's

polychotomous, that is nominal scale at

more than just two levels,

ordinal scale outcome variables.

You can have other situations

that are not simply binary.

But the, but the most commonly understood

and used logistic regression models is for

the situation where you have

a binary outcome variable.

And the techniques that we use in linear

regression that most people are familiar

with actually provide the motivation for

the approach that we took for

logistic regression in our books.

course on Applied Logistic Regression.

This is an eight week long course.

And we are going to be going

through Approximately four of

the chapters in the textbook that

the course is based on and named after,

called Applied Logistic Regression.

Where I am one of the authors,

along with my colleagues,

David Hosmer and Rodney Sturdivant.

For those of you who are interested

in the book, you may find

the book is available, I think,

anywhere you can get books.

Again, the title is

Applied Logistic Regression.

The third edition was released in

March of 2014, March of 2013.

So, it's relatively new, and again,

we will be covering perhaps four or

so chapters of this book.

If you get interested

in logistic regression,

I strongly recommend that at some point

you get the book and you can find

many more topics beyond what we'll be

able to cover together in this course.

So we we are going to do a lot together.

Certainly, the lectures are going to take,

are going to take us time to go through.

But you really will learn this course,

you'll really learn to apply

logistic regression models

by doing the homeworks that you'll

find at the end of each of the weeks.

And I've given you, and

you'll find on the website,

you'll find the the,

the homework assignments.

And you will also find solutions

to the homework assignments.

And, it's, the expectation is that you

will download the statistical package

STATA, into your computers, and that you

can carry out the assignments using STATA.

And it will make it easier if you

use STATA because I've given you all

the solutions as well.

So the idea here is not to have

you become a STATA expert,

but the idea is to give

you practical experience,

running logistic regression models on,

with the computer program.

And interpreting the output

of those programs.

And you can do that very easily

with STATA with this course.

But if you prefer to use SPSS,

or SASS, or R, or

anything else that you're used to,

you can certainly do that.

The material in this course is such that

most of what we do is available in

all modern statistical packages.

So you have the option to do

that yourself if you would like.

So, we'll start right away with

logistic regression analysis, and

the goal of a logistic regression

analysis is to find the best-fitting,

simplest model that we possibly can.

That describes the relationship between

an outcome or dependent variable or

response variable and a set of independent

variables which we call predictors or

explanatory variables.

We also call them covariates.

But what distinguishes a logistic

regression model from a linear regression

model, is that the outcome variable

in linear regression is binary or

dichotomous.

Now actually, if you did get our book and

you looked at some of the later chapters

of the book, you realize that,

that's not even a 100% true statement.

Because in the more advanced

logistic regression modeling,

you may have an outcome variable that's

polychotomous, that is nominal scale at

more than just two levels,

ordinal scale outcome variables.

You can have other situations

that are not simply binary.

But the, but the most commonly understood

and used logistic regression models is for

the situation where you have

a binary outcome variable.

And the techniques that we use in linear

regression that most people are familiar

with actually provide the motivation for

the approach that we took for

logistic regression in our books.

Загрузка...

Выбрать следующее задание

Ты добавил

Выбрать следующее задание

Ты добавил