Question

QUESTION 1)

What do we need for translating the probability of categorical outcome to class membership?

Group of answer choices:

a) The logit.

b) Hyperparameters.

c) The odds ratio.

d) A cutoff value.

QUESTION 2)

Which of the following is true regarding profiling and classification using logistic regression?

A) The goal of profiling is to identify the significant predictors that help differentiate between class 1 and class 0.

B) The goal of classification is predicting which class an observation would belong to, based on the values of predictor variables.

Group of answer choices:

a) Only B

b) Both A and B

c) Only A

d) Neither A nor B

QUESTION 3)

What would happen if we use this Python statement:

             data_df = pd.get_dummies(data_df, drop_first=False)

as a data preprocessing step in a linear model, regardless of regression or classification?

Group of answer choices:

a) Underfitting of the test data because of too few variables

b) Perfect collinearity for a 2-class categorical variable or perfect multicollinearity for a multi-class categorical variable, which will destabilize the model.

c) Overfitting of the training data because of too many variables.

d) Perfect collinearity for a 2-class categorical variable or perfect multicollinearity for a multi-class categorical variable, but the model will run okay.

QUESTION 4)

Which of the following statements is true with regard to interpreting the output of logistic regression?

A) A greater-than-1 value of beta coefficient indicates that a higher value on the predictor is associated with a higher probability of belonging to class 1

B) A less-than-1 value of beta coefficient indicates that a higher value on the predictor is associated with a lower probability of belonging to class 1

Group of answer choices:

a) Only B

b) Both A and B

c) Neither A nor B

d) Only A

Thank you so much for the help!

Answer 1

1)

Logit a type of function that creates a map of probability values from (0,1) - infinity to +infinity

That is categoral value to class.

hyperparameter is a parameter whose value is used to control the learning process.

odds ratio (OR) is a measure of association between an exposure and an outcome. The OR represents the odds that an outcome will occur given a particular exposure, compared to the odds of the outcome occurring in the absence of that exposure.

When we use cutoff when dealing with unbalanced datasets

So option A is correct.

2)

The statements A and B both related to classification only.

Profiling related to time complexities and program analysis.

Classification describes the class which it should belong.

So only b is correct.

Option A.

3)the line indicates,

Convert categorical variable into dummy/indicator variables.

data_df----array-like, Series, or DataFrame
Data of which to get dummy indicators.

drop_firstbool, default False
Whether to get k-1 dummies out of k categorical levels by removing the first level.

Perfect collinearity for a 2-class categorical variable or perfect multicollinearity for a multi-class categorical variable, but the model will run okay.

Option D

4)yes,

A greater-than-1 value of beta coefficient indicates that a higher value on the predictor is associated with a higher probability of belonging to class 1
A less-than-1 value of beta coefficient indicates that a higher value on the predictor is associated with a lower probability of belonging to class 1

Option B.

Please do upvote thank you.