Monthly Subscription Starting at AED 99 + VAT

Bayesian Statistics: Time Series Analysis
  • Level Professional
  • المدة 22 ساعات hours
  • الطبع بواسطة University of California, Santa Cruz
  • Offered by

عن

This course for practicing and aspiring data scientists and statisticians. It is the fourth of a four-course sequence introducing the fundamentals of Bayesian statistics. It builds on the course Bayesian Statistics: From Concept to Data Analysis, Techniques and Models, and Mixture models. Time series analysis is concerned with modeling the dependency among elements of a sequence of temporally related variables. To succeed in this course, you should be familiar with calculus-based probability, the principles of maximum likelihood estimation, and Bayesian inference. You will learn how to build models that can describe temporal dependencies and how to perform Bayesian inference and forecasting for the models. You will apply what you've learned with the open-source, freely available software R with sample databases. Your instructor Raquel Prado will take you from basic concepts for modeling temporally dependent data to implementation of specific classes of models

الوحدات

Introduction

    • 1 Videos
    • 2 Readings
    • 1 Assignment
Show info about module content

1 Assignment

  • Objectives of the course

1 Videos

  • Welcome to Bayesian Statistics: Time Series

2 Readings

  • Introduction to R
  • List of References

Stationarity, the ACF and the PACF

    • 3 Videos
    • 4 Readings
    • 1 Assignment
Show info about module content

1 Assignment

  • Stationarity, the ACF and the PACF

3 Videos

  • Stationarity
  • The autocorrelation function (ACF)
  • ACF, PACF, Differencing and Smoothing: Examples

4 Readings

  • The partial autocorrelation function (PACF)
  • Differencing and Smoothing
  • R Code: Differencing and filtering via moving averages
  • R Code: Simulate data from a white noise process

The AR(1) process: Definition and properties

    • 2 Videos
    • 2 Readings
    • 1 Assignment
Show info about module content

1 Assignment

  • The AR(1) definitions and properties

2 Videos

  • The AR(1)
  • Simulating from an AR(1) process

2 Readings

  • The PACF of the AR(1) process
  • R Code: Sample data from AR(1) processes

The AR(1): Maximum likelihood estimation and Bayesian inference

    • 3 Videos
    • 4 Readings
    • 1 PeerReview
    • 1 Assignment
Show info about module content

1 Assignment

  • MLE and Bayesian inference in the AR(1)

1 Peer Review

  • MLE and Bayesian inference in the AR(1)

3 Videos

  • Maximum likelihood estimation in the AR(1)
  • Bayesian inference in the AR(1)
  • Bayesian inference in the AR(1): Conditional likelihood example

4 Readings

  • Review of maximum likelihood and Bayesian inference in regression
  • R code: MLE for the AR(1), examples
  • R Code: AR(1) Bayesian inference, conditional likelihood example
  • Bayesian inference in the AR(1), full likelihood example

The general AR(p) process

    • 4 Videos
    • 3 Readings
    • 1 Assignment
Show info about module content

1 Assignment

  • Properties of AR processes

4 Videos

  • Definition and state-space representation
  • Examples
  • ACF of the AR(p)
  • Simulating data from an AR(p)

3 Readings

  • Rcode: Computing the roots of the AR polynomial
  • Rcode: Simulating data from an AR(p)
  • The AR(p): Review

Bayesian inference in the AR(p)

    • 5 Videos
    • 5 Readings
    • 1 PeerReview
    • 1 Assignment
Show info about module content

1 Assignment

  • Spectral representation of the AR(p)

1 Peer Review

  • Bayesian analysis of an EEG dataset using an AR(p)

5 Videos

  • Bayesian inference in the AR(p): Reference prior, conditional likelihood
  • Model order selection
  • Example: Bayesian inference in the AR(p), conditional likelihood
  • Spectral representation of the AR(p)
  • Spectral representation of the AR(p): Example

5 Readings

  • Rcode: Maximum likelihood estimation, AR(p), conditional likelihood
  • Rcode: Bayesian inference, AR(p), conditional likelihood
  • Rcode: Model order selection
  • Rcode: Spectral density of AR(p)
  • ARIMA processes

The Normal Dynamic Linear Model: Definition, model classes, and the superposition principle

    • 4 Videos
    • 2 Readings
    • 1 Assignment
Show info about module content

1 Assignment

  • The Normal Dynamic Linear Model

4 Videos

  • NDLM: Definition
  • Polynomial trend models
  • Regression models
  • The superposition principle

2 Readings

  • Summary of polynomial trend and regression models
  • Superposition principle: General case

Bayesian inference in the NDLM: Part I

    • 6 Videos
    • 5 Readings
    • 1 PeerReview
Show info about module content

1 Peer Review

  • NDLM: sensitivity to the model parameters

6 Videos

  • Filtering
  • Filtering in the NDLM: Example
  • Smoothing and forecasting
  • Smoothing in the NDLM: Example
  • Second order polynomial: Filtering and smoothing example
  • Using the dlm package in R

5 Readings

  • Summary of the filtering distributions
  • Rcode. Filtering in the NDLM: Example
  • Summary of the smoothing and forecasting distributions
  • Rcode: Smoothing in the NDLM, Example
  • Rcode: Using the dlm package in R

Testing branch

    • 1 Assignment
Show info about module content

1 Assignment

  • NDLM, Part I: Review

Seasonal NDLMs

    • 2 Videos
    • 2 Readings
    • 1 Assignment
Show info about module content

1 Assignment

  • Seasonal Models and Superposition

2 Videos

  • Fourier representation
  • Building NDLMs with multiple components: Examples

2 Readings

  • Fourier Representation: Example 1
  • Summary: DLM Fourier representation

Bayesian inference in the NDLM: Part II

    • 3 Videos
    • 2 Readings
    • 1 PeerReview
Show info about module content

1 Peer Review

  • NDLM data analysis

3 Videos

  • Filtering, Smoothing and Forecasting: Unknown observational variance
  • Specifying the system covariance matrix via discount factors
  • NDLM, Unknown Observational Variance: Example

2 Readings

  • Summary of Filtering, Smoothing and Forecasting Distributions, NDLM unknown observational variance
  • Rcode: NDLM, Unknown Observational Variance Example

Case studies

    • 2 Videos
Show info about module content

2 Videos

  • EEG data
  • Google trends

Testing branch

    • 1 Assignment
Show info about module content

1 Assignment

  • NDLM, Part II

Data Analysis Project

    • 1 PeerReview
Show info about module content

1 Peer Review

  • Data Analysis Project

Auto Summary

Bayesian Statistics: Time Series Analysis, offered by Coursera and taught by Raquel Prado, is designed for data scientists and statisticians. This advanced course, the fourth in a series, focuses on modeling temporal dependencies and performing Bayesian inference using R. A solid understanding of calculus-based probability and Bayesian principles is required. With a duration of 1320 minutes, it offers both Starter and Professional subscription options. Ideal for professionals looking to deepen their expertise in time series analysis.

Raquel Prado
Instructor

Raquel Prado