Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Chemistry > Biostats test 4 > Flashcards

Biostats test 4 Flashcards

1
Q

What is binary logistic regression

A

prediction of a binary-valued DV on the basis of other variables, so response variable is not continuous, but binary-valued, as in:

yes/no, has/does not have, alive/dead, increased/decreased

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

values of outcome variable

A

failure (coded 0) or success (coded 1)

success means “has the property”

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

what do we try to predict with binary logistic regression

A

the probability of success as a function of covariates

p(success) = f(cov1, cov2, …)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Assumptions of binary logistic

A
  • categories must be mutually exclusive (no overlap) and collectively exhaustive (all cases can be assigned)
  • if so, for all cases, success or failure can be coded in the data
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

in logistic regression, for predicted probabilities to be meaningful, they must…..

A

lie between the values of 0 and 1

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Assumptions of logistic regression

A

categories must be mutually exclusive (no overlap) and collectively exhaustive (all categories can be assigned)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Link function

A

transforms the dependent variable so outcome range can be restricted to be between 0 and 1

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Similarities of BLR with OLS

A
  • model building and its issues (colinearity, order of entry, influential cases)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Dissimilarities of BLR with OLS

A
  • DV is binary (categorical), not continuous
  • Interpretation of coefficients
  • Assessment of model fit/quality of obtained model
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

How do we ensure the [0,1] restircuted outcome range for the predicted values

A
  • link function (logit) is used to relate the linear model part to the outcome variable
  • it transforms the predicted values so that the outcomes are restrained to fall in the meaningful 0 to 1 range
  • Regression techniques that make use of some kind of link function are called Generalized Linear Models (GLM)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

The logit function

A
  • Natural logarithm of odds
  • Logit = ln(odds) = ln(y hat/1-y hat)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

The logistic regression model when combined with the logit (link) function

A

ln(y hat/1-y hat) = b0 + b1X1 + b2X2 + …

So the difference with OLS model is on the dependent variable side of the model

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

In Logistic Regression, rather than looking at the B coefficients themsleves, we look at

A
  • ODDS RATIO = e to the bower of b, where e is the base of natural log ln
  • change in the odds (of success) for a one-unit change in the predictor
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Wald test

A

Gives the p-value of the odds ratio

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Under H0, odds ration is

A
  • Odds ratio = 1
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What if we want to calculate the odds ratio for a non-unit size

A

multiply regression coefficient by the size before you raise e to the power of the coefficient

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Multiplicative effect

A

The combined effect of predictors on the DV is a product of separate effects, so the effects of odds ratios multiply in binary logistic regression, while they add up in ordinary linear regression

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

From probabilities to classification - what is classification of cases based on?

A

The predicted probabilities for success. The default setting for classification as ‘success’ is a predicted probability for success > .50.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

confusion matrix

A

the confusion matrix in your output gives predicted versus observed successes and failures, based on the cut-value (always think of a 2x2 table).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Classification errors

A

false positives and false negatives

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

The null model

A

starting point in the model building, a model without any actual predictors

if have to predict without knowing anything about predictors, the LARGEST CATEGORY WINS!

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

How to decide which covariates to include

A

in the end, need to classify cases as success or fail cases - two groups

Suppose you have a continuous covariate and success and fail groups differ on mean → covariate may help to classify as success or fail

Suppose you have categorical covariates, and success and fail groups have different distributions. Chi-square test of homogeneity will be significant → categorical covariate may help to classify as success or fail

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Receiver operating characteristic curve

A

a plot that illustrates the performance of a binary classifier system for different discrimination methods

typically gives true positive rate (specificity) against the false positive (1 - sensitivity) rate at various cut value settings

allows to see how diff cut value settings affect that classification results of your classifier

lenient threshold (low cut value): sensitivity is up!

strict threshold (high cut value): specificity is up!

Optimum typically: combination of high sensitivity and high specificiity

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

A lenient (low) cut value leads to

A

higher sensitivity! Success easily detected (at the expense of increased false positives)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

A strict (high) cut values leads to

A

higher specificity! Good at keeping false positives down, but less sensitive

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

ROC curve

A

area under curve is overall indicator of diagnostic accuracyw

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

what does (P/1-P) in the logistic regression model represent

A

It represents the ODDS of the event happening

28
Q

what data does logistic regression model use to calculate probabilities, eg in the height and gender example and how does it do it

A

Data it uses:
- predictor variable (x)
- binary dependent vairble

How it uses it:

  1. during training the model fits the data to the logit = B0 + B1X equation, and learns B0 and B1 by estimating
    how X changes with Y
  2. uses predictor variables to calculate log odds (by inputting the predictor into the logistic regression equation), then converts them into probabilities using log transformations
  3. uses the cutoff to classify the observation into one of the two outcomes
29
Q

odds ratios

A

change in the odds (of success) for a one-unit change in the predictor

30
Q

under H0, what is the odds ratio for logistic regression model

A

1

31
Q

survival analysis is about

A

analyzing time-to-event data

32
Q

examples of descriptive research questions that would warrant survivial data

A

median survival time in clinical studies

how does the probability for an event change over time?

33
Q

examples of inferential research questions that would warrant survivial data

A
  • What variables explain the time to event best?
  • Do they shorten or lengthen the expected time?
  • By how much?
34
Q

mortality

A

refers to dropout in data

35
Q

censored data

A

when you have partial information: if a case drops out, you do not know when if ever, the endpoint was reached. survival analysis is developed to take censored data into account.

36
Q

right-censored

A

event not experienced at the end of the study - might or might not occur later, we don’t know

37
Q

left censored

A

event occufrs during study, but starting point lies before start of study, so exact time interval is not known

38
Q

interval censored

A

event occurs during study, but not exactly known when

39
Q

non-censored data

A

start time is known, end point of interest reached during study

40
Q

non-informative sensoring

A

censoring is not related to the likelihood of developing the event of interest

41
Q

hazard rate

A

the probability for event to occur in a next (small) time interval, assuming the event has not yet occured. like instantaneous risk

42
Q

survival rate

A

cumulative probability for non-event for a certain amount of time after some starting point

43
Q

Mortality

A

Subjects are lost to follow-up (drop out) before the end of the study so you do not know if/when the end point was ever reached. this is one of the reasons for not being able to use logistic regression for time-to-event data.

44
Q

censored data

A

incomplete time to event data

45
Q

right censored cases

A

event not experienced at the end of the study
It might or might not occur later –we don’t know

could be dropout too

46
Q

left censored cases

A

event occurs during study, but starting point lies before start of study, exact time interval not known

47
Q

interval censored cases

A

event occurs during study, but not exactly known when

47
Q

non-censored data

A

start time is known, end point of interest reached during study

48
Q

non-informative censoring

A

when in survival regression, we assume that censoring is not related to the likelihood of developing the event of interest and that that subjects whose data are censored would have the same distribution of times to event, had they actually been observed

49
Q

hazard rate

A

instantaneous risk: the probability that if case survived to time t, event will be experienced in the next time interval t + Δt

50
Q

survival rate

A

cumulative probability for the non-event for a certain amount of time after some starting

51
Q

median survival time

A

time to event on average

52
Q

Describing survival data using life tables

A
  • Break down range of survival times into smaller time slices
  • Tabulate the counts of all relevant events (including censored data) per time slice
  • Probabilities for ‘at risk’, of ‘dying’, and of ‘surviving’ can be computed on the basis of these counts
  • look at cumulative survival data per time slice
53
Q

Kaplan-Meier life table

A
  • every case has its own row in data file
  • For each case, information about the survival time and censoring is entered
  • The resulting curves are more smooth than for grouped life tables, as time is specified per subject, not per (fixed) time slice
54
Q

How to get around confounders playing a role in survival regression

A

If confounders play a role, life tables can present a misleading picture.
However, you can create survival curves for different categories and compare these.

55
Q

what is the log rank test used for

A

comparing survival times for different groups

56
Q

how does the log rank test work

A

it computes scaled difference between observed and expected number of events per time slice, which are then combined

57
Q

what is cox proportional hazard regression used for

A

determining whether there is a significant relationship between one or more covariates and hazard, quantifying and testing these relationships, and generating a prognosis curve

58
Q

dependent variable of the cox proportional hazard regression

A

ln(hazard)

59
Q

prognosis curve

A

a predicted survival curve customized for any specific combination of covariate values

60
Q

Building cox regression model

A
  1. baseline curve of mean values for all covariates
  2. model coefficients determined (if greater than 1, hazard is up and survival is down)
61
Q

what is h0 in

ln (h) = ln(h0) + b1X1 + b2X2 + … bkXk


A

baseline hazard - overall shape of curve

62
Q

calculating h from ln (h) = ln(h0) + b1X1 + b2X2 + … bkXk


A

h = h0 x e^(b1X1 + b2X2 + ….)

So hazard is baseline hazard multiplied with exponential of model

63
Q

hazard ratio (relative risk) explanation and interpretation

A

Exp(b) or e to the power of b

relative risk of experiencing the event (e.g., failure, death) between two groups or for a one-unit increase in a predictor.

64
Q

Proportional hazard assumption

A

Effects of covariates on the likelihood for the event are assumed constant over time, so ratio of hazards is constant (proportional) over time

Chemistry (16 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2025 Bold Learning Solutions. Terms and Conditions