Statistical Inference & Regression Models Flashcards
7 cards from real MS-DS Master of Data science practice questions. Tap to flip, then mark Knew It or Still Learning — missed cards come back until you master them.
Read the first 7 Statistical Inference & Regression Models flashcards as text
Which test is most appropriate for comparing the means of three or more independent groups?
Answer: One-way ANOVA F-test
One-way ANOVA tests whether at least one group mean differs from the others by comparing between-group and within-group variance.
In ridge regression, the regularization parameter λ controls:
Answer: The degree of shrinkage applied to regression coefficients
Ridge regression adds an L2 penalty (λΣβ²) to the loss function, shrinking coefficients toward zero as λ increases.
A Type II error in hypothesis testing occurs when:
Answer: A false null hypothesis is not rejected
Type II error (false negative) is failing to reject H₀ when H₀ is actually false; its probability is denoted β.
The coefficient of determination R² in a simple linear regression equals 0.64. What does this mean?
Answer: 64% of the variance in Y is explained by X
R² measures the proportion of total variance in the dependent variable explained by the regression model.
In a Poisson regression model, the canonical link function is:
Answer: The log link
Poisson regression uses the natural log as its canonical link function, modeling log(μ) as a linear combination of predictors.
The power of a statistical test is defined as:
Answer: The probability of rejecting H₀ when H₁ is true
Power = 1 - β = P(reject H₀ | H₁ is true), representing the test's ability to detect a true effect.
In a simple linear regression, which diagnostic plot is most useful for detecting non-linearity?
Answer: Residuals vs. fitted values plot
A residuals vs. fitted values plot reveals non-random patterns that indicate non-linearity in the relationship between predictors and response.