Concept

Parametric model

Related lectures (32)

Linear Estimation & Prediction: Models & Methods

Explores linear estimation and prediction in AR parametric models, focusing on Yule Walker equations and Wiener filter.

Explores copulas, margins, extremes, and parametric models for higher dimensions and bivariate data.

Linear Independence: Covariance and Statistical Dependence

Explores covariance, statistical dependence, education-fertility relationship, hypothesis testing, and comparison statistics for continuous outcomes.

Statistical Estimation: Gaussian Linear Model

Delves into statistical estimation, highlighting the Gaussian linear model and the limitations of ML estimators.

Important Sampling: Monte Carlo Estimation

Covers important sampling for efficient Monte Carlo estimation of expected values using a new distribution to reduce variance.

Hypothesis Testing in Statistics

Delves into hypothesis testing, covering test statistics, critical regions, power functions, p-values, multiple testing, and non-parametric statistics.

Probability Distributions in Environmental Studies

Explores probability distributions for random variables in air pollution and climate change studies, covering descriptive and inferential statistics.

Extreme Region Likelihood Estimation

Covers extreme region likelihood estimation, model complexity, oceanographic data modeling, and threshold likelihood estimation.

Surface Integrals: Implicit Surfaces

Covers implicit surfaces, parametric descriptions, regular parametrization, normal vectors, and orientable surfaces.

Survival Analysis: Cox Proportional Hazards Model

Introduces the Cox proportional hazards model for survival analysis, covering estimation, assessment of assumptions, and interpretation using residuals.

Statistical Estimation: Models and Learning Machines

Covers statistical estimation models, ML estimators, learning machines, practical issues, and challenges in estimation.

Block Models: Continued Analysis

Explores the stochastic blockmodel, spectral clustering, and non-parametric understanding of blockmodels, emphasizing metrics for comparing graph models.