Logistic Regression

Dimension reduction

is genuine in that the reduced sets of selected variables are actual inputs rather than transformations of the inputs.

significant improvement

The forward selection terminates when no ________ can be made.

Variable Clustering

finds groups of inputs that are as correlated as possible among themselves and as uncorrelated as possible with variables in other clusters.

Odds Ratio

shows the strength of the association between the predictor variable and the response variable.

Logistic Regression

________ is very similar to linear regression.

Estimation methods

provide customized imputations for each case with missing values by training a model to predict an inputs value from other inputs.

Regression Analysis

Helps one understand how the typical value of the dependent variable changes when any one of the independent variables is varied, while the other independent variables are held fixed.

Logistic Regression

________ is used to predict the categorical dependent variable using a given set of independent variables.

Regression Analysis

Helps one understand how the typical value of the dependent variable changes when any one of the independent variables is varied, while the other independent variables are held fixed

What is the focus of regression

The focus is on the relationship between a dependent variable (i.e., Response, outcome, target) and one or more independent variables (or 'predictors‘, ‘features)

What is understood better with regression

Helps one understand how the typical value of the dependent variable changes when any one of the independent variables is varied, while the other independent variables are held fixed

Synthetic distribution methods

replacing all missing values of a variable with a fixed value

Estimation methods

provide customized imputations for each case with missing values by training a model to predict an inputs value from other inputs

Variable Redundancy

A consequence of input redundancy is model instability

Variable Clustering

 finds groups of inputs that are as correlated as possible among themselves and as uncorrelated as possible with variables in other clusters

Dimension reduction

is genuine in that the reduced sets of selected variables are actual inputs rather than transformations of the inputs

Exhaustive search

you try every combination

Restrictive search

an alternative to the exhaustive search is to restrict the search to a sequence of improving models

Forward selection

creates a sequence of models of increasing complexity

Odds Ratio

shows the strength of the association between the predictor variable and the response variable

Accuracy

(TRUE POSITIVES AND NEGATIVES) / (TOTAL CASES)

TP + TN / (TP + TN + FP + FN)

Error Rate

(FALSE POSITIVES AND NEGATIVES) / (TOTAL CASES)

FP + TN /(TP+TN+FP+FN)

Sensitivity(Recall)

(TRUE POSITIVES) / (TOTAL ACTUAL POSITIVES)

TP /(TP + FN)

Precision(Positive Predictive Value)

(TRUE POSITIVES) / (TOTAL PREDICTED POSITIVES)

TP/(TP+FP)

Specificity

(TRUE NEGATIVES) / (TOTAL ACTUAL NEGATIVES)

TN / (TN+FP)

What is logistic regression used to predict?

values of categorical variables

What type of problems is Logistic Regression used for to solve?

Classification

By finding the S-curve what are we able to accomplish?

Classifying samples

What must the output be for Logistic Regression?

Categorical variables

What is not required of Logistic Regression?

A linear relationship between the dependent and independent variables.

Logit Link function

a logit is defined as the log base e (log) of the odds p/(1-p) where p is the probability.

Logit Link function equation

log(p-hat/1-[p-hat]) = ^B0 + ^B1x1+^B2x2

p-hat from logit link function equation

1/1+e^-logit(^p)