by Albert Major
SIMPLE LINEAR REGRESSION
Why this matters for CSAI:
• Regression is one of the core tools in data science and machine learning.
• You will use this logic again for model evaluation, predicting outcomes, feature importance, and understanding variance.
• Later topics like multiple regression, interactions, classification models, and even neural networks build on this idea of “predicting Y from X”.
CORE IDEAS
• Regression is about predicting an outcome Y from a predictor X.
• You fit a line to data points.
• The line gives you a systematic rule to predict Y for any value of X.
• You check how good that line is.
• You test if the line improves prediction over a trivial baseline (the mean of Y).
THE MODEL
You model a straight line.
Y = b0 + b1X + error
• X is the predictor.
• b0 is the intercept. It is the value of Y when X is zero.