Skip to the content.

Forecasting and Time Series Methods

This is a course in the analysis of time series data with emphasis on appropriate choice of models for estimation, testing, and forecasting. Topics or methodologies covered may include Univariate Box-Jenkins for fitting and forecasting time series; ARIMA models, stationarity and nonstationarity; diagnosing time series models; transformations; forecasting: point and interval forecasts; seasonal time series models; modeling volatility with ARCH, GARCH; modeling time series with trends; and other methods. The R Shiny App development will be covered to help students obtain skills of making prototype of their models and ideas.

Contributors:
Dr. Yichen Qin, Associate Professor of Business Analytics at University of Cincinnati Dr. Martin Levy, Professor of Business Analytics at University of Cincinnati

Lecture

Introduction to time series, R, Rmarkdown

Intro

Stationarity, strictly stationary, weakly stationary, backshift operator, white noise ACF, PACF

Stationarity, ACF, PACF

Autoregression & Moving Average

AR, MA

ARMA & ARIMA

ARMA, ARIMA

Model fitting

Model fitting

Augmented Dickey-Fuller test (ADF)

ADF

Forecast

Forecast

Seasonal ARIMA

Seasonal ARIMA

vector autoregressive models (VAR)

VAR