STATS 4A03/6A03, Winter 2019
Time series analysis is concerned with data consisting of time-ordered sequences of measurements on some phenomenon of interest. These type of data are common in many areas, including business (weakly interest rates, daily closing stock prices), climate (daily high and low temperatures, annual amount of precipitation), agriculture (annual crop and livestock production figures, annual export sales), public consumption (hourly provincial hydro demand, yearly garbage amount produced by a city), and many others. Unlike most other statistical data, time series data show correlation over time, we call it autocorrelation. In addition, the data often show trends such as linear, polynomial, seasonal and sinusoidal patterns. Upon successful completion of this course, the students will be able to perform the following tasks:
(a) produce, interpret and explain time plots, including identification of trends, for a time series dataset;
(b) calculate, plot and interpret autocorrelation functions;
(c) fit and interpret time series models such as ARMA (autoregressive moving average) and ARIMA (autoregressive integrated moving average) to data; (d) carry out forecasts from time series data and interpret and explain the results; and
(e) Handle the numerical calculations and plotting in the computer using statistical software such as R.
INSTRUCTOR: R. Viveros
Stationary, auto-regressive and moving-average series, Box-Jenkins methods, trend and seasonal effects, tests for white noise, estimation and forecasting methods, introduction to time series in the frequency domain.
Three lectures; one term
Prerequisite(s): STATS 3A03, STATS 3D03
PLEASE REFER TO MOSAIC FOR THE MOST UP-TO-DATE INFORMATION ON TIMES AND ROOMS