Logistic Regression and Survival Analysis- Austin

studied byStudied by 33 people
5.0(1)
Get a hint
Hint

How does a logistic distribution compare to normal distribution?

1 / 25

encourage image

There's no tags or description

Looks like no one added any tags here yet for you.

26 Terms

1

How does a logistic distribution compare to normal distribution?

  • look similar BUT shape on the edges are different

    • aka homoscedasticity

<ul><li><p>look similar BUT shape on the edges are different</p><ul><li><p>aka homoscedasticity</p></li></ul></li></ul>
New cards
2

When the dependent variable is continuous we use linear regression, and when the dependent variable is discrete we use _________________.

logistic regression

New cards
3

What are the pros and cons of clinical categories?

pros: quickly make distinction bc we have cutoffs

cons: can be misleading

New cards
4

For binary logistic regression it is the appropriate method when an outcome variable consists of only ____ categories.

2 (yes/no)

New cards
5

Probabilities must be constrained between ___ and ____ in binary logistic regression.

0 and 1

New cards
6

What is logit or “log-odds” ?

the natural log of the odds of the event of interest occurring

New cards
7

In logistic regression we cannot use OLS like multiple regression, so instead we use what?

MLE

New cards
8

In MLE we are looking at the ____________ ability for a variable to effect an outcome.

maximum

New cards
9

In logistic regression, we cannot use R2 like multiple regression, instead we use what?

chi-square

New cards
10

Is binary logistic regression nonparametric or parametric?

nonparametric

New cards
11

In binary logistic regression, in the confidence interval if it crosses what number is the variable insignificant?

1

New cards
12

What does it mean when a regressor has a coefficient estimate that has a negative sign?

variable decreases probability

New cards
13
<p>What is this equation used for?</p>

What is this equation used for?

Binary logistic regression

New cards
14

When is survival analysis used?

when the research question is WHEN something will occur… not IF

New cards
15

Survival Analysis is used a lot in what setting in medicine? FYI

cancer treatments

New cards
16

In logistic regression, the coefficients are hard to interpret, so instead what do we look at?

odds ratio

New cards
17

What are the 3 components of a survival analysis?

  1. Kaplan-Meier Method/Curve

  2. A Cox Proportional Hazard Ratio

  3. log-rank test that produces a p-value

New cards
18

The Kaplan-Meier method and cox proportional hazard ratio all MUST account for the possibility of what?

censoring

New cards
19

What does censoring/ censored data mean in this context?

This occurs when the time to event occurrence is not precisely known for individuals

  • BASICALLY: lack of data

  • Ex: If a patient drops out of a study, we don’t know what happened to them? did they die? survive? etc.

New cards
20

The Kaplan-Meier Method summarizes what?

the probability of survival over time estimated from a sample

New cards
21

How is hypothesis testing done in the Kaplan-Meier Method?

logrank test

New cards
22

What is a hazard?

risk of event occurrence at time t

New cards
23

A hazard function is hazard for an event over _______.

time

New cards
24
<p>This is the equation for a hazard function. What would I do to that equation to create the cox proportional hazards regression model?</p>

This is the equation for a hazard function. What would I do to that equation to create the cox proportional hazards regression model?

Just take the natural log (ln)!!!

<p>Just take the natural log (ln)!!!</p>
New cards
25

In Cox Proportional Hazards Regression Model, what happens is B>0?

an increase in the predictor leads to an increase in the hazard (or decrease in survival)

  • Ex: cigarettes smoked a day. If I INCREASE the predictor (# of cigarettes), that leads to an increase in hazard (lung cancer)

New cards
26

In Cox Proportional Hazards Regression Model, what happens is B<0?

an increase in the predictor leads to an decrease in the hazard (or increase in survival)

  • Ex: exercise

New cards

Explore top notes

note Note
studied byStudied by 60 people
Updated ... ago
5.0 Stars(3)
note Note
studied byStudied by 5 people
Updated ... ago
5.0 Stars(1)
note Note
studied byStudied by 34 people
Updated ... ago
5.0 Stars(1)
note Note
studied byStudied by 18 people
Updated ... ago
5.0 Stars(1)
note Note
studied byStudied by 8 people
Updated ... ago
4.5 Stars(2)
note Note
studied byStudied by 31 people
Updated ... ago
5.0 Stars(2)
note Note
studied byStudied by 225 people
Updated ... ago
5.0 Stars(1)
note Note
studied byStudied by 38 people
Updated ... ago
5.0 Stars(1)

Explore top flashcards

flashcards Flashcard46 terms
studied byStudied by 7 people
Updated ... ago
5.0 Stars(1)
flashcards Flashcard89 terms
studied byStudied by 24 people
Updated ... ago
5.0 Stars(1)
flashcards Flashcard46 terms
studied byStudied by 2 people
Updated ... ago
5.0 Stars(1)
flashcards Flashcard36 terms
studied byStudied by 4 people
Updated ... ago
5.0 Stars(1)
flashcards Flashcard47 terms
studied byStudied by 16 people
Updated ... ago
5.0 Stars(3)
flashcards Flashcard20 terms
studied byStudied by 12 people
Updated ... ago
5.0 Stars(1)
flashcards Flashcard67 terms
studied byStudied by 177 people
Updated ... ago
5.0 Stars(4)
flashcards Flashcard31 terms
studied byStudied by 3 people
Updated ... ago
5.0 Stars(1)