Time Series ARIMA Models

Time series ARIMA models are applied with time series data of variables measured over time. Time series analysis examines relationships of variables over time such as commodity prices or crop yields. Time series models may be used for analyzing the effects of a specific event (such as the effects of the recession on unemployment rates) or for forecasting (for example to predict economic growth or future prices).

Handouts, Programs, and Data

Time Series ARIMA Models

Time Series ARIMA Models Example

Time Series ARIMA Models Stata Program and Output

Time Series ARIMA Models in Stata.do

Time Series ARIMA Simulations.do

timeseries_ppi.dta

Time Series ARIMA Models R Program and Output

Time Series ARIMA Models in R.R

timeseries_ppi.csv

Time Series ARIMA Models

SAS Program and Output

Time Series ARIMA Models in SAS.sas

timeseries_ppi.csv

Time series models: topics covered

• • White noise, autoregressive (AR) models, moving average (MA) models, ARMA models

• Stationarity, differencing, detrending, seasonality

  • Dickey-Fuller test for stationarity

  • Autocorrelation function (ACF) and partial autocorrelation function (PACF)

  • Box-Jenkins methodology for selecting an ARIMA model