Linear regression interpretation, logistic regression, overfitting, validation and model metrics.
42 cards · basic cards · AI-written, checked twice. Edit anything.
- What does a regression coefficient measure?
Show answer
The change in the dependent variable for a one-unit increase in the independent variable, holding others constant- What is the intercept in a regression equation?
Show answer
The predicted value of the dependent variable when all independent variables equal zero- What does R-squared measure?
Show answer
The proportion of variance in the dependent variable explained by the independent variables- What is a residual?
Show answer
The difference between the observed value and the predicted value for a data point- How do correlation and regression differ?
Show answer
Correlation measures the strength of association between two variables; regression predicts one variable from another- What does adjusted R-squared account for?
Show answer
The number of predictors in the model, penalizing for adding variables that do not improve fit- What is multicollinearity?
Show answer
High correlation among independent variables, which inflates standard errors and reduces reliability of coefficients- What are the five key assumptions of linear regression?
Show answer
Linearity, independence of errors, homoscedasticity, normality of residuals, and no multicollinearity- What is heteroscedasticity?
Show answer
Non-constant variance of residuals across different levels of the independent variable- What is autocorrelation?
Show answer
Correlation of residuals with their own past values, common in time-series data- What is overfitting?
Show answer
A model that learns the training data too well, including its noise, resulting in poor generalization to new data- How does overfitting affect performance metrics?
Show answer
Training metrics appear excellent while test metrics degrade, showing a large gap between them- What is underfitting?
Show answer
A model that is too simple to capture the underlying pattern in the data- What is train-test split?
Show answer
Dividing data into two subsets to train the model on one and evaluate it on the other- What is k-fold cross-validation?
Show answer
Dividing data into k subsets, training k models each leaving out one fold, then averaging performance metrics
FlashKeepers