1What is the predicted variable in multiple linear regression (MLR)?
A numerical (continuous) value.
2What are the two main objectives of MLR?
To establish if there is a relationship between input variables and the output and to predict new observations.
3How does MLR describe the relationship between variables?
By fitting a line to the observed data to minimize the overall modeling error.
4What does the equation of MLR include?
A dependent variable modeled as a function of several independent variables with corresponding coefficients.
5What is the role of regression coefficients in MLR?
They represent the increase/decrease factor that a change in an independent variable has on the predicted value of the dependent variable.
6What does R-squared represent in MLR?
It measures how well the model accounts for variations in predictions explained by predictor variables.
7When is a predictor variable considered statistically significant?
When the p-value is less than 0.05.
8What is the method used to find the best-fitting line in MLR?
The method of least squares.
9What does a high R-squared value indicate?
That the model has a strong linear association and explains most of the variation in the dependent variable.
10What is the impact of adding more predictor variables to a model?
It increases the complexity and may improve the model’s predictive accuracy.
11What does the least squares approach help with in MLR?
It helps estimate the model coefficients to minimize errors and fit the line of best fit.
12What is the main objective of simple linear regression?
To establish if there is a statistically significant relationship between an input variable and an output variable, and to predict new observations.
13What type of values does the predicted variable in simple linear regression have?
Numerical or continuous values.
14What does simple linear regression aim to minimize?
The overall modeling error by fitting a line to the observed data.
15What is the independent variable usually denoted as?
X.
16What is the dependent variable usually denoted as?
Y.
17What is the purpose of the error term in simple linear regression?
It is the residual between the observed value and the predicted value that lies on the regression line.
18What is the general form of the linear equation in simple linear regression?
y = B₀ + B₁X + e, where B₀ is the y-intercept, B₁ is the slope, X is the independent variable, and e is the error term.
19How do you assess if a predictor variable is statistically significant?
By checking if the p-value is less than 0.05.
20What does an R-square value indicate?
It measures how much of the variability in the output variable is explained by the input variable.
21What is the adjusted R-square value?
It accounts for the proportion of variation explained by the predictor, adjusted for the number of predictors used in the model.
22What was the predicted salary for a candidate with 5 years of experience in the given example?
K73,545.90.
23What does the slope in a simple linear regression model represent?
The change in the dependent variable for every one-unit increase in the independent variable.
24What is the significance of the constant in a linear regression model?
It is the y-intercept, representing the predicted value when the independent variable is zero.
25What is the p-value used for in a simple linear regression model?
To test the statistical significance of the predictors or the overall model.
26What happens when the adjusted R-square value is close to 1?
It indicates that the model explains a high proportion of the variability in the dependent variable.
27What does the line-of-best-fit do in simple linear regression?
It minimizes the total error between the observed values and the predicted values.
28What is logistic regression used for?
Classification purposes and predicting the probability of a target variable
29What type of learning is logistic regression?
Supervised machine learning algorithm
30What are the possible outcomes in logistic regression?
Dichotomous outcomes, such as 0 or 1, Yes or No, True or False
31What shape does the logistic regression function take?
An "S" shaped function
32What is the difference between logistic and linear regression?
Logistic regression predicts probabilities and is used for classification, while linear regression predicts continuous outcomes
33What are the types of dependent variables in logistic regression?
Binomial (binary), multinomial, or ordinal
34What is an example of a binomial outcome?
True or False, Yes or No, Pass or Fail
35What does multi-collinearity mean in logistic regression?
A high correlation between two predictor variables that can lead to unreliable models
36What is support vector machine (SVM)?
A supervised ML algorithm used for classification and regression modeling
37What is the main objective of SVM?
To find an optimal hyperplane that separates data points into different output classes
38How does SVM handle outliers?
It is not sensitive to outliers but involves some penalties
39What is an example of SVM usage?
Image recognition, such as predicting whether an image is of a cat or a dog
40What are the pros of SVM?
High accuracy, works well with high-dimensional space, uses less memory
41What are the cons of SVM?
Consumes more time in training/testing, not suitable for large datasets