DSC 12-Time Series Analysis Guidelines

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 3

GUIDELINES FOR DISCIPLINE-SPECIFIC CORE COURSE –12:

TIME SERIES ANALYSIS

Credits: 4
Number of classes per week: 5 (3 Lecture + 2 Practical)

Reference Books:
[1]. Chatfield, C. and Haipeng Xing (2019): The Analysis of Time Series Introduction to R, 7th
edition, CRC Press
[2]. Mukhopadhyay, P. (1999): Applied Statistics, 2nd Ed, Reprint 2015, New Central Book Agency.
[3]. Gun, A.M., Gupta, M.K. and Dasgupta, B. (2008): Fundamentals of Statistics, Vol. II, 9th Ed
[4]. Gupta, S.C., and Kapoor, V.K. (2008): Fundamentals of Applied Statistics, 4th ed (reprint 2010),
Sultan Chand and Sons.
[5]. Montgomery, D.C., Jennings, C. L. and Kulachi, M. (2015): Introduction to Time Series and
Forecasting, 2nd ed, Wiley Series in Probability and Statistics.
[6]. Kendall, M. (1976): Time Series, 2nd Ed, Charles Griffin and Company Ltd.
[7]. Montgomery, D.C. and Johnson, L.A. (1976): Forecasting and Time Series Analysis, McGraw
Hill Book Company.

UNIT TOPIC BK; CH; SEC./ PG NO.;


SUB-SEC.;
UNIT I: Introduction to Times series data and [1]; 1; 1.1; 1-8
Time Series its applications. [2]; 14; 14.1; 397-399
Data and its Components of a time series and its [1]; 2; 2.1; 15-17
Components decomposition. [2]; 14; 14.2; 399-401
[4]; 2; 2.2-2.3; 2.3-2.7
Estimation of trend component by all [1]; 2; 2.5.1-2.5.2; 20-25
four methods and effect of elimination [2]; 14; 14.3; 402-408
of trend on other components of the [4]; 2; 2.4; 2.7-2.9,
Time-series. 2.11-2.40
[2]; 14; 14.5; 410-411
[3]; 7; 7.9; 434-436
Estimation of the seasonal component [4]; 2; 2.5; 2.41-2.50
under fixed and changing seasonal [3]; 7; 7.6; 425-429
patterns, De-seasonalization of data. [4]; 2; 2.5.5; 2.51-2.52
UNIT II: Simple sinusoidal model; [1]; 7; 7.1, 7.2; 122-126
Spectral Periodogram, and Harmonic Analysis. [1]; 7; 7.3; 126-130
Analysis [3]; 7; 7.7; 429-434
and Variate-difference method. [4]; 2; 2.9; 2.67-2.69
Stationarity Time series, and Stochastic process; [1], 2; 2.2; 17
Stationarity. [2]; 15; 15.2; 447,
[2]; 15; 15.3.2; 453-454
Autocorrelation; meaning, definition, [4]; 2; 2.8; 2.60-2.62
causes, consequence, and test for [1]; 4; 4.7; 109-111
autocorrelation.
UNIT III: Stochastic Models: White noise [1]; 3; 3.4, 3.5; 45-47
Time Series Process, Random walk.
Models MA, AR, ARMA, and their properties [1]; 3; 3.6, 3.7; 47-62
using correlogram, ACF, and PACF, [4]; 2; 2.7; 2.57-2.59
Yule walker equations, Fitting of [1]; 4; 4.2, 4.3, 81-94
AR(1), AR(2), MA(1), MA(2), and 4.4;
ARMA(1,1) process. [1]; 3; 72-74
(All problems
except problems
3.10 & 3.14)
Non-stationary models: ARIMA and [1]; 3; 3.9; 63-64
SARIMA models. [1]; 4; 4.8; 103-104
[5]; 5; 5.6, 5.6.1,
5.9;
Dicky Fuller test, Augmented Dickey- [1]; 4; 4.6; 99-102
Fuller test.
Wold’s Decomposition Theorem. [1]; 3; 3.13; 70-72
Non-linear time series models: ARCH [5]; 7; 7.3; 355-357
and GARCH Process. (507-512)
UNIT IV: Principles of Forecasting; [1]; 5; 5.1; 115-117
Univariate Performance Evaluation. [7]; 1; 1.2; 3-4
Forecasting Extrapolation of Trend Curves. [1]; 5; 5.2.1; 118
Procedures Exponential smoothing; Holt- [1]; 5; 5.2.2, 118-122
Winter’s. 5.2.3;
Box- Jenkins’ Methodology. [1]; 5; 5.2.4; 123-127
[1]; 5; 146-147
(Problems 5.1-
5.6 in the
exercise)

NOTE: Teachers are advised to encourage students to take up group projects (optional) by using
secondary data on various topics covered in the syllabus.

PRACTICAL / LAB WORK

List of Practical:

1. Fitting and plotting of modified exponential curves by different methods.


2. Fitting and plotting of Gompertz curve by different methods.
3. Fitting and plotting of logistic curves by different methods.
4. Fitting of the trend by the Moving Average Method for a given extent and for an estimated extent.
5. Measurement of Seasonal indices: a) Fixed and b) Changing Patterns
6. Construction of Periodogram and Harmonic Analysis
7. Estimation of variance of the random component
8. Construction of Correlogram for given AR(1), AR(2), MA(1), MA(2), and ARMA(1,1) processes.
9. Fitting of AR(1), AR(2), MA(1), MA(2), and ARMA(1,1) processes for given datasets.
10. Forecasting by various exponential smoothing procedures.
11. Forecasting by Box-Jenkins methodology.
(May be done using EXCEL/SPSS/R/Calculators)
PATTERN OF THE QUESTION PAPERS
Total Marks: 160 (Theory-120, Practical-40)

Theory
Duration: 3 hrs.
Max. Marks: 120 (End Semester Exam- 90, IA-30)
1. The question paper will consist of seven questions, out of which students are to attempt any five
questions.
2. The seven questions will be set such that:
There will be two questions from UNIT I, one from UNIT II, two from UNIT III, and one from
UNIT IV, so the weightage given to units I and III are 36 marks each, and that to units II and IV is 18
marks each.
Practical
Duration:3 hrs.
Max Marks: 40 (End Semester Exam.-20, Viva-10, CA-10)
The question paper will contain five/four questions. Students are to attempt any four/three questions,
depending on the usage of the computer/calculator and the size of the datasets given in the questions.

You might also like