Regression & Time Series

Prerequisites

• SVI 201 Statistical Inference

Learning Objectives

• Explain how 202 extends 201's single-quantity inference to relationships (regression) and to time (time series).
• Fit a simple linear regression by ordinary least squares and interpret the slope and intercept in the units of the problem.
• Write the multiple-regression model in matrix form and compute the OLS estimator β̂ = (XᵀX)⁻¹XᵀY.
• Read residuals as the window into the OLS assumptions, and explain why heteroskedasticity and autocorrelation leave the slope unbiased but make its standard errors wrong.
• Define weak stationarity and connect non-stationary prices to a unit root — the random walk is an AR(1) with φ = 1.
• Explain volatility clustering (magnitude, not direction) and why it breaks the constant-variance assumption.
• Explain Monte Carlo as estimating an expectation by averaging simulations, justified by the law of large numbers, with error shrinking like 1/√N.

Module Schedule