Details of MA5110 (Autumn 2019)

Level: 5 Type: Theory Credits: 4.0

Course CodeCourse NameInstructor(s)
MA5110 Time series analysis Satyaki Mazumder

Introduction: Review of various components of time series, plots and descriptive statistics, discrete-parameter stochastic processes-- strong and weak stationarity, autocovariance and autocorrelation.

Spectral Analysis and Different Processes: Spectral analysis of weakly stationary processes-- periodogram, fast Fourier transform; Moving average, autoregressive, autoregressive moving average (ARMA) and autoregressive integrated moving average processes (ARIMA); Box-Jenkins model, state-space model.

Forecasting and Model Selection: Linear filters, signal processing through filters, inference in ARMA and ARIMA models; Forecasting-- ARIMA and state-space models, Kalman filter; Model building-- residuals and diagnostic checking; Model selection-- strategies for missing data.

Time-frequency Analysis: Short-term Fourier transform, wavelets, data analysis with computer packages.

Statistical Inference (MA4107) and Functional Analysis (MA4102).


Suggested Texts:

  1. Brockwell, P.J. and Davis, R.A., Introduction to Time Series and Forecasting, Springer.

  2. Fuller, W.A., Introduction to Statistical Time Series, Wiley-Blackwell.

  3. Shumway, R.H. and Stoffer, D.S., Time Series Analysis and Its Applications, Springer.

Course Credit Options

Sl. No.ProgrammeSemester NoCourse Choice
1 IP 1 Not Allowed
2 IP 3 Not Allowed
3 IP 5 Not Allowed
4 MR 1 Not Allowed
5 MR 3 Not Allowed
6 MS 9 Elective
7 RS 1 Not Allowed
8 RS 2 Not Allowed