INDIAN STATISTICAL INSTITUTE

POST-GRADUATE DIPLOMA
IN
STATISTICAL METHODS AND ANALYTICS

STUDENT BROCHURE
Effective from the Academic Year 2021-2022
 POST-GRADUATE DIPLOMA IN STATISTICAL
METHODS AND ANALYTICS

Contents
I. GENERAL INFORMATION 1

1. Scope 1
2. Eligibility 1
3. Stipend and Contingency Grant 1
4. Selection Procedure 2
5. Course Structure 2
6. Examinations and Scores 2
7. Satisfactory Conduct 5
8. Promotion 7
9. Final Result 7
10. Award of Certificates 8
11. Prizes and Medals 8
12. Class-Teacher 8
13. Attendance 8
14. Stipend 9
15. Library Rules 10
16. Hostel Facilities 10
17. Change of Rules 10

II. CURRICULUM 11

18. Courses for Semester I 11

19. Courses for Semester II 16
 POST-GRADUATE DIPLOMA IN STATISTICAL
METHODS AND ANALYTICS

This one year (two semester) programme is offered at Chennai Centre, Chen-
nai and North-East Centre, Tezpur. Fifty percent seats of North-East Centre
are reserved for the North-East domicile students.

I. GENERAL INFORMATION

1 Scope
The course is intended to provide students with a comprehensive and rig-
orous training in basic theory and applications of Statistical Methods and
Analytics, in addition to some exposure to Mathematics and Computational
Techniques. The students are exposed to data handling using R and Python
packages. The programme is so designed that on successful completion, the
students will be able to take up jobs in industries and government depart-
ments where applications of Statistics and Data Analytics are required.

2 Eligibility
In order to be eligible for admission to this programme, a candidate must
have a three-year Bachelor’s degree in any discipline with Mathematics as one
of the subjects or a B.E./[Link]. degree or any other qualification considered
equivalent (such as A.M.I.E. diploma).
The North-East domicile students must provide a domicile certificate of
North-Eastern states from a recognized authority.
Any student who is asked to discontinue the programme is not eligible for
readmission in to this programme.

3 Stipend and Contingency Grant

There is no tuition fee. The North-East domicile students admitted to this
programme will receive a monthly stipend of Rs 3000/- for a period of eleven
months and an annual contingency/book grant of Rs 3000/-. In the first

1
instance, stipends will be granted for the first semester only and renewed
if the progress of the student is found to be satisfactory. Stipend granted
to a student may be reduced or fully withdrawn if the academic progress,
attendance in class or character and conduct of the student are not found
satisfactory (Further details in Section 14).
There is no stipend for the non-domicile students at Chennai and North-East
Centres.

4 Selection Procedure
Selection is based on the performances in written test and interview. Past
academic records may also be taken into consideration. The written test
will comprise of multiple-choice questions in Mathematics and Probability at
pass/minor level of Bachelor’s degree.

5 Course Structure
The one-year programme consists of a total of 10 courses. There are five
courses in the first semester and four courses plus a one-course project in the
second semester.

Semester I Semester II

1. Probability Theory 6. Statistical Methods II

2. Statistical Methods I 7. Statistical Machine Learning
3. Statistical Inference 8. Statistical Modeling
4. Vectors & Matrices and 9. Time Series and
Regression Methods Statistical Finance
5. Programming: R and Python 10. Project

The project work is likely to extend through summer.

All students from North-East Centre may be required to spend one week at
the headquarters of the Institute (Kolkata) at the end of Semester I.

6 Examinations and Scores

Courses 1, 2, 3, 5, 6, 7 and 8 are Non-Module-Based courses. Courses 4
and 9 are Module-Based courses, each consisting of two modules.

2
Non-Module-Based Courses
The final (semester) examination in a course is held at the end of the semester.
Besides, there is a mid-semester examination in each course. The calendar
for the semester is announced in advance. The composite score in a course
is a weighted average of the scores in the mid-semester and semester exam-
inations, class tests, homework, assignments, and/or project work in that
course. The weights are announced beforehand by the Dean of Studies, or
the Students’ Academic Affairs In-Charge or the Class Teacher in consul-
tation with the concerned teacher or the concerned teacher. The minimum
composite score to pass a course is 35%.

Module-Based Courses
There will be one examination at the end of the module for each of the two
modules for any Module-Based course. Equal weight will be given to the two
modules for computing the composite score of the course. The score in a
module is a weighted average of the scores in the internal assessments and
the end of the module examination. For each module, the weights to be given
to homework, assignments, class tests, end-module examination etc will be
announced beforehand by the Dean of Studies or the Students’ Academic
Affairs In-Charge or the Class Teacher in consultation with the concerned
teacher or the concerned teacher. The minimum composite score to pass a
course is 35%.

Back-Paper Examination
For both types of courses, if the composite score of a student falls short of
45% in a course, the student may take a back-paper examination to improve
the score. At most one back-paper examination is allowed in each course.
Moreover, a student can take at most two back-paper examinations in the
first semester and at most one in the second semester. The decision to
allow a student to appear for the back-paper examination is taken by the
appropriate Teachers’ Committee. The back-paper examination covers the
entire syllabus of the course.
In case of back-paper examination in a module-based course, there would be
one single question paper covering both the modules with equal distribution
of marks over the two modules. The total score obtained in a back-paper
examination of any module-based course would be the total of the marks
obtained in the two modules.
When a student takes a back-paper examination in any of these two types
of courses, his/her final score in that course is the higher of the back-paper
score and the earlier composite score, subject to a maximum of 45%.

3
A student may take more than the allotted quota of back-paper examinations
in a given academic year and decide at the end of the academic year which
of the back-paper examination scores should be disregarded.

Compensatory Paper Examination

A student who gets less than 35% in at most one course even after the
back-paper examination in any semester, but 60% or more in average in the
other courses in that semester is allowed to appear for a compensatory pa-
per examination. In case of a module-based course, there will be one single
question paper, like the back-paper examination, covering both the modules
with equal distribution of marks over the two modules. A student will be al-
lowed to appear in at most one compensatory paper in the entire programme.
Maximum marks obtainable in a compensatory paper will be 35%. In the
second semester, a student will have to choose between the compensatory
paper examination and the possibility of repeating the programme. He/she
will not be allowed to take both.
A student will have to discontinue the programme if he/she scores less than
35% in the compensatory paper in any semester.

Supplementary Examination
If a student misses the mid-semester or semester examination of a course
or the examination for a module of a module-based course due to medical
or family emergency, the Teachers’ Committee may allow him/her to take a
supplementary examination in the course for the missed examination, based
on an adequately documented representation from the student. The supple-
mentary semester examination for a non-module-based course is held at the
same time as the back-paper examination for the semester and the student
taking the supplementary semester examination in a course is not allowed
to take any further back paper examination in that course. For a module-
based course, the supplementary examination is held at a convenient time.
The maximum that a student can score in a supplementary examination is
60%. Unlike the back-paper examination, the score in the supplementary
examination is used along with other scores to arrive at the composite score.
There will be supplementary examination for mid-semester, semester, back-
paper and compensatory examinations within a month of the examination
missed by a student due to medical or family emergency. The student should
submit a written application to the Dean of Studies or the Academic Af-
fairs In-Charge for appearing in the supplementary examination, enclosing
supporting documents. On receipt of such an application from a student
with supporting documents, the Dean of Studies or the Academic Affairs

4
In-Charge,, in consultation with the relevant Teachers’ Committee, will de-
cide whether such examination will be allowed. The student can score at
most 60% in the supplementary examinations of mid-semester and semester
examinations. For the back-paper or the compensatory papers, the maxi-
mum the student can score in the supplementary examination is 45% or 35%
respectively.

Project
The one-course project in second semester will be on data analytics and will
be supervised by a faculty. There is a regular assessment of the project work
during the semester and at the end of the semester. The evaluation at the
end of the semester is based on the final project report and presentation. The
schedule and type of regular assessments and the weights to be given to the
regular assessments and end-semester evaluation are announced beforehand
by the Dean of Studies, or the Students’ Academic Affairs In-Charge or the
Class Teacher.

7 Satisfactory Conduct
A student is also required to maintain satisfactory conduct as a necessary con-
dition for taking semester examination, for promotion and award of diploma.
Failing to follow the examination guidelines, copying in examination, rowdy-
ism, other breach of discipline of the Institute, unlawful/unethical behaviour
and the like are regarded as unsatisfactory conduct. Violation of such na-
ture is likely to attract punishments such as withholding promotion/award
of diploma, withdrawing stipend and/or expulsion from the hostel/Institute.
Ragging is banned in the Institute. If any incident of ragging comes to
the notice of the authorities, the concerned student shall be given liberty to
explain, and if his/her explanation is not found to be satisfactory, he/she may
be expelled from the Institute. The punishment may also take the shape of

1. suspension from the Institute for a limited period,

2. suspension from the classes for a limited period,
3. withholding stipend/fellowship or other benefits,
4. withholding results,
5. suspension or expulsion from hostel and the likes.

Local laws governing ragging are also applicable to the students of the Insti-
tute.

5
Guidelines during Examinations
The students are required to abide by the following guidelines during the
examinations.

i) Students are required to take their seats according to the seating ar-
rangement displayed. If any student takes a seat not allotted to him/her,
he/she may be asked by the invigilator to hand over the answer script
(i.e., discontinue the examination) and leave the examination hall.

ii) Students are not allowed to carry inside the examination hall any mo-
bile phone with them – even in a switched-off mode. Calculators, books
and notes will be allowed inside the examination hall only if these are so
allowed by the teacher(s) concerned (i.e., the teacher(s) of the course),
or if the question paper is an open-note/open-book one. Even in such
cases, these articles cannot be shared.

iii) No student is allowed to leave the examination hall without permission

from the invigilator(s). Further, students cannot leave the examination
hall during the first 30 minutes of any examination. Under no circum-
stances, two or more students writing the same paper can go outside
together.

iv) Students should ensure that the main answer booklet and any extra
loose sheet bear the signature of the invigilator with date. Any dis-
crepancy should be brought to the notice of the invigilator immedi-
ately. Presence of any unsigned or undated sheet in the answer script
will render it (i.e., the unsigned or undated sheet) to be cancelled, and
this may lead to charges of violation of the examination rules.

v) Any student caught cheating or violating examination rules will get

‘Zero’ in that examination. If the offence is in a back-paper examina-
tion, the student will get ‘Zero’ in the back-paper. (The other condi-
tions for promotion, as mentioned in Section 8 below, will continue to
hold).

The decisions regarding promotion in Section 8 and final result in Section 9

below are arrived at after taking in to account the violation, if any, of the
satisfactory conduct by the student, as described in this section.

6
8 Promotion
A student is considered for promotion to the second semester of the pro-
gramme only when his/her conduct has been satisfactory. Subject to the
above condition, a student is promoted from first semester to second semester if

i) the number of composite scores less than 45% is at most two and
ii) no composite score in a course is less than 35%.

Otherwise, a student is not promoted to the second semester and he/she is

asked to discontinue the programme.

9 Final Result
At the end of the second semester, the overall average of the percentage
composite scores in all the courses taken in the two-semester programme
is computed for each student. The student is awarded the post-graduate
diploma in one of the following categories according to the criteria he/she
satisfies provided in the second semester,

i) he/she does not have a composite score of less than 35% in any course,
ii) the number of scores less than 45% is at most one and
iii) his/her conduct is satisfactory.

Post-Graduate Diploma in Statistical Methods with Applications: passed in

First Division with Distinction if

i) the overall average score is at least 75%, and

ii) the composite score in at most one course is less than 45%.

Post-Graduate Diploma in Statistical Methods with Applications: passed with

First Division if

i) the overall average score is at least 60%,

ii) the composite score in at most one course is less than 45%, and
iii) not obtained First Division with Distinction.

Post-Graduate Diploma in Statistical Methods with Applications: passed with

Second Division if

i) the overall average score is at least 45%,

ii) the composite score in at most two courses is less than 45%, and
iii) not obtained First Division with Distinction or First Division.

7
All others students are considered to have failed. A student who fails but
obtains at least 35% average score in the second semester, and have satis-
factory conduct is allowed to repeat the programme without any stipend all
throughout the year provided that he/she has not taken the option of a com-
pensatory paper examination in the second semester. A student is not given
more than one chance to repeat.

10 Award of Certificate
A student passing the Diploma is given a certificate which includes

i) the list of all courses taken along with the respective composite scores,
and
ii) the category (Passed with Distinction or Passed) of his/her final result.

The certificate is awarded in the Annual Convocation of the Institute follow-

ing the semester II examinations.

11 Prizes and Medals

Students may be awarded prizes in form of book awards for good academic
performances in each semester, as decided by the Teachers’ Committee.

12 Class Teacher
One of the instructors of a class is designated as the Class Teacher. Students
are required to meet their Class Teacher periodically to get their academic
performance reviewed and to discuss their problems regarding courses.

13 Attendance
Every student is expected to attend all the classes. If he/she is absent, he/she
must apply for leave to the Dean of Studies or the Academic Coordinator.
Failing to do so may result in disciplinary action.

8
14 Stipend
Stipend, if awarded at the time of admission, is valid initially for the first
semester only. The amount of stipend to be awarded in the second semester
will depend on academic performance and conduct, as specified below, pro-
vided the requirements for continuation of the academic programme (exclud-
ing repetition) are satisfied.
Performance in course work: The composite scores considered for the fol-
lowing performance criteria are the composite scores after the respective
back-paper examinations

i) If all the requirements for continuation of the programme are sat-

isfied, and the average composite score is at least 60% and the
number of courses with scores less than 45% is at most two in the
first semester, then the full value of the stipend is awarded in the
second semester.
ii) If all the requirements for continuation of the programme are sat-
isfied, and the average composite score is at least 45% and the
number of courses with scores less than 45% is at most one in the
first semester, then the half value of the stipend is awarded in the
second semester.
iii) In all other cases no stipend is awarded in the second semester.

Attendance: If the overall attendance in all courses in the first semester is

less than 75%, no stipend is awarded in the following semester.

Conduct: At any time, the Dean of Studies or the Academic Affairs In-
Charge or the Class Teacher, in consultation with the respective Teach-
ers’ Committee, may withdraw the stipend of a student fully for a
specific period if his/her conduct in the campus is found to be unsat-
isfactory.
The net amount of the stipend to be awarded is determined by simultaneous
and concurrent application of all clauses described above but in no case the
amount of stipend to be awarded or to be withdrawn should exceed 100% of
the prescribed amount of stipend. Stipends are given after the end of each
month for eleven months in the academic year. The first stipend is given two
months after the admission with retrospective effect provided the student
continues in the Diploma programme for at least two months. Contingency
grants can be used for purchasing a scientific calculator and other required
accessories for the practical class, text books and supplementary text books

9
and for getting Photostat copies of required academic material. All such
expenditure should be approved by the Class Teacher. No contingency grants
are given in the first two months after admission.

15 Library Rules
A student is allowed to use the reading room facilities in the library and
allowed access to the stacks. Student will have to pay a security deposit
in order to avail himself/herself of the borrowing facility. The amount of
security deposit will be specified at the time of the admission to the course.
A student can borrow at most four books at a time. Fine is charged if any
book is not returned by the due date stamped on the issue-slip. The library
rules and other details are available in the library of the Centre to which the
student is admitted.

16 Hostel Facilities
In case the Hostel Accommodation is provided to the students, the students
will have to pay the caution money and the monthly room rent.

17 Change of Rules
The Institute reserves the right to make changes in the above rules, course
structure and the curriculum as and when needed.

10
II. CURRICULUM

Semester I Semester II

1. Probability Theory 6. Statistical Methods II

2. Statistical Methods I 7. Statistical Machine Learning
3. Statistical Inference 8. Statistical Modeling
4. Vectors & Matrices and 9. Time Series and
Regression Methods Statistical Finance
5. Programming: R and Python 10. Project

18 Courses for Semester I

1. Probability Theory

Elementary concepts of probability: experiments, outcomes, sample

space, events, axiomatic definition of probability and its properties. (8)
Conditional probability, independence, Bayes theorem. (5)
Random variable, probability distribution and properties; probability
mass/density function, cumulative distribution function, expectation,
variance, mean square error, moments. (7)
Bernoulli and Binomial, Poisson, Geometric and Negative Binomial,
Hypergeometric, Uniform, Normal Exponential, Gamma, Beta distri-
butions. (8)
Chebyshev’s inequality, weak law of large numbers (statement and con-
cept), central limit theorem (statement and concept). (2)
Distribution of a function of a random variable. (4)
Bivariate distribution for discrete and continuous random variables;
joint, marginal and conditional distributions, moments, covariance, cor-
relation coefficient. Bivariate Normal Distribution (8)
Independent random variables and their sums. Transformation of two
random variables. (6)
Sampling distributions under the assumption of normality: chi-square,
t, F. (4)

11
 Suggested References:
a) Introduction to Probability and Statistics for Engineers and Sci-
entists - Ross, S.
b) Elementary Probability Theory - Chung, K. L.
c) Introduction to Probability Theory - Hoel P. G., Port S. C. and
Stone, C. J.
d) Introduction to Probability - Roussas, G.
e) Applied Statistics and Probability for Engineers - Montgomery,
D.C. and Runger, G.C.
f) A First Course in Probability - Ross, S.

2. Statistical Methods I

Introduction to data, statistical problems and related data analysis.

Concept of population, sample and statistical inference through exam-
ples. (4)
Summarization of univariate data; graphical methods, measures of lo-
cation, spread, skewness and kurtosis; outliers and robust measures,
sample moments. Empirical cumulative distribution function. (12)
Statistical computations: data summary and graphical display of data,
basic statistics. Plotting empirical cumulative distribution function. (4)
Data simulations from discrete and continuous probability distribu-
tions: Bernoulli and Binomial, Poisson, Geometric and Negative Bi-
nomial, Hypergeometric, Uniform, Normal, Exponential, Chi-Square,
Gamma, Beta. (8)
Analysis of discrete and continuous data: fitting some standard discrete
and continuous probability distributions. Goodness of fit: Pearson’s
Chi-square test, Kolmogorov-Smirnov test. Graphical methods of veri-
fying the fit: Q-Q and P-P Plots. Shapiro-Wilks test for Normality. (8)
Introduction to resampling (bootstrap) and cross-validation techniques.
(4)
Introducing the analysis of variance; one way analysis, F test, Kruskal-
Wallis nonparametric test; two way analysis. Designs of experiment:
principles of designing an experiment. Introduction to CRD, RBD,
Balanced and Unbalanced Block Designs, cross over designs with ap-
plications in industrial and clinical trials. (12)
(All computations and data analysis are expected to be done using
R/Python.)

12
 Suggested References:
a) Introductory Statistics - Ross, S
b) Statistics - Freedman, D., Pisani R. and Purves, R.
c) Applied Statistics and Probability for Engineers - Montgomery,
D.C. and Runger, G.C.
d) Introduction to Probability and Statistics for Engineers and Sci-
entists - Ross, S.
e) Design and Analysis of Experiments - Montgomery, D.C.
f) An Introduction to Statistics with Python- Haslwanter T.
g) A Handbook of Statistical Analysis using R - Everitt, B. S. and
Hothorn, T.

3. Statistical Inference

Brief introduction to random variable and probability distributions.

Random sample and the concept of statistical Inference with exam-
ples.(6)
Point estimation: estimator and estimate. Desirable properties of an
estimator: unbiasedness, smaller variance and mean squared error.
Method of moments and maximum likelihood estimation. Asymptotic
behaviour of MLE (statement on consistency and asymptotic normal-
ity). (8)
Interval estimation- Confidence interval and its basic properties, Con-
struction of confidence interval for parameter of Uniform distribution,
exponential distribution, mean of the normal distribution with known
and unknown variance. (6)
Hypotheses and the concept of hypotheses testing. Null and alterna-
tive, simple and composite hypothesis, significance level, size, p-value
and power. Introduction to likelihood ratio tests with examples. (12)
One sample problem: Test for randomness - run test. Test for mean
under the assumption of normality with known and unknown variance.
Nonparametric tests for median: signed test, Wilcoxon’s signed rank
test. One sample test for proportion. (10)
Comparison of two samples- Two independent samples: graphical pro-
cedures, K-S test. Comparing mean under the assumption of normality
(two sample t test). Nonparametric test for medians - Mann-Whitney-
Wilcoxon. Two sample test for proportion. Two dependent samples:
paired t test under the assumption of normality, Nonparametric tests
for two dependent samples. (10)

13
 Suggested References:
a) Introductory Statistics - Ross, S
b) Introduction to Statistical Theory - Hoel P. G., Port S. C. and
Stone, C. J.
c) An Introduction to Probability and Statistical Inference - Roussas,
G.
d) Applied Statistics and Probability for Engineers - Montgomery,
D.C. and Runger, G.C.
e) Introduction to Probability and Statistics for Engineers and Sci-
entists - Ross, S.

4. Vectors & Matrices and Regression Methods

Vectors & Matrices (24)

Introduction to Vectors and matrices.

Vectors: Definition and examples, vector spaces and subspaces, basis
of a vector space, linear dependence/independence (4).
Matrices: Definition and examples, matrix as a linear transformation,
elementary matrices and elementary matrix operations, basic matrix
operations including those of partitioned matrices, rank, nullity, trace,
determinant and inverse of a matrix, idempotent matrix and its prop-
erties. Solutions of system of equations (14)
Spectral theory: eigenvalues and eigenvectors of matrices, decomposi-
tion of matrices, quadratic forms and definiteness of a matrix (with
applications in Statistics) (6)

Suggested References:
a) Matrix Algebra: Theory, Computations and Applications in Statis-
tics - Gentle, J. E.
b) Matrix Algebra useful for Statistics - Searle, S.
c) Matrices with Applications in Statistics - Graybill, F. A.
d) Matrix Algebra From a Statistician’s Perspective - Harville, D. A.

Regression Methods (28)

Introduction to Classical Linear Regression Model (2).

OLS method of estimation; fitted values, prediction of the response
variable, tests of hypotheses. (6)

14
 Residuals. Validation of assumptions using graphical techniques. Re-
gression Diagnostics (10)
Use of dummy variables in regression (2);
Variable selection, multicollinearity. Model selection using AIC and
BIC criteria. (4)
Concepts of robust and nonparametric regression (4).
Illustration of the methodology with real data.

Suggested References:
a) Introduction to Linear Regression Analysis - Montgomery, D. C.,
Peck, E. and Vinning, G.
b) Regression Analysis by Examples - Chatterjee S. and Hadi, G.
c) Applied Linear Regression - Weisberg, S.
d) Applied Regression Analysis - Draper, N.R. and Smith, H.

5. Programming: R and Python

Introduction to packages- R and Python: overview of packages, data

handling, input-output operations. Basic programming: data types,
arrays, loops etc.; functions and graphics.(12 + 12)
Improvement of the initial solution using methods of bisection, sort and
search, Regula Falsi and Newton-Raphson. (6)
Fixed point iterative schemes, significant digits, round-off errors, finite
computational processes and computational errors. Order of conver-
gence and degree of precision.(4)
Matrix computations - basic operations, finding determinant, inverse,
eigen roots and eigen vectors of a matrix, matrix decomposition, solving
system of equations. (6)
Computational aspects of constrained optimization (4).
Unconstrained optimization: Newton, Quasi-Newton method. (4)
Experimentation designs: Obtaining global optimal solutions from lo-
cal optimum solutions using iterative experimentation like Response
Surface Methodology. (4)

Suggested References:
a) Beginning R: The Statistical Programming Language - Gardener,
M.

15
 b) Computational Statistics: An Introduction to R - Sawitzki, G
c) Optimization - Lange, K.
d) Dive into Python - Pilgrim, M.
e) An Introduction to Statistics with Python- Haslwanter T.
f) A Handbook of Statistical Analysis using R - Everitt, B. S. and
Hothorn, T.

19 Courses for Semester II

6. Statistical Methods II

Multivariate Data Exploration/visualization, multivariate data han-

dling, random vector, mean and variance-covariance matrix, introduc-
tion to multinomial and multivariate normal distributions. (10)
Applied Multivariate techniques: Principal components analysis and
Factor Analysis. (10)
Introduction to discrete time Markov chains, finite and countable state
space. Introduction to Markov chain Monte Carlo (MCMC) methods
and applications of MCMC method in statistics. Introduction to HMM
algorithm. (10)
Handling missing data; various methods of imputations including hot
deck algorithm (MICE in R), EM algorithm (6)
Advanced regression techniques: Ridge, Principal component regres-
sion, LASSO and Spline smoothing (8)
Multiple Hypotheses testing. (4)
Introduction to MANOVA and Hotelling’s t2 . (4)
Illustration of the methodology with real data.

Suggested References:
a) Applied Multivariate Statistical Analysis - Johnson, R.A. and
Wichern, D.W
b) A User’s Guide to Business Analytics - Basu, A. and Basu, S.
c) Computational Statistics - Gentle, J. E.
d) Handbook of Computational Statistics - Gentle, J.E., Hardle, W.
and Mori, Y.

16
 e) Statistical Computing: Existing Methods and Recent Develop-
ments - Basu, A. and Kundu, D.
f) Applied Nonparametric Regression - Hardle, W.
g) Introducing Monte Carlo Methods with R - Robert, C. P. and
Casella, G.

7. Statistical Machine Learning

Introduction to bootstrap based machine learning. Assessment and

model selection: confusion matrix and various criteria of evaluation,
training and testing error rates. (8)
Pattern Recognition and Classification techniques
Unsupervised learning: clustering procedures: hierarchical and non-
hierarchical, k-means; association rules, ROCs. (8)
Supervised learning: Linear and quadratic discriminant analysis; Bayesian
classifier, nearest neighbour classifier, Entropy based classifier. (10)
Tree based classification methods: predictive modeling using decision
trees (CART), random forests. (8)
Support vector machine. Introduction to boosting and adaptive boost-
ing algorithm. (6)
Introduction to Natural Language Processing (NLP), information re-
trieval and text analysis: stop words, TF-IDF measure, vector space
models. (6)
Introduction to neural networks, Convolutional NN, Deep NN. (6)
Illustration of the methodology with real data.

Suggested References:
a) The Elements of Statistical Learning With Applications in R -
James, G., Witten, D., Hastie, T. and Tibshirani, R.
b) Introduction to Statistical Learning Theory - Hastie, T., Tibshi-
rani, R. and Friedman, J.
c) A User’s Guide to Business Analytics - Basu, A. and Basu, S.
d) Data Mining: Concepts and Techniques - Han, J. and Kamber,
M.
e) Classification and Regression Trees - Breiman, L. et al
f) Statistical and Machine-Learning Data Mining - Ratner, B.

17
8. Statistical Modeling

Logistic regression; odds ratio, concordance-discordance measures, Lo-

gistic Regression as a classsifier. Probit Regression. Introduction to
Multilogit models. (10)
Modeling count data: Poisson Regression, Poisson models for zero in-
flated data. (8)
Introduction and visualizing categorical data. Measures of association.
Loglinear Models, Models for nominal and ordinal response. (8)
Survival Data Modeling: Time-to-event data and survival probabilities,
notion of censoring, survival curve and other ways of representing sur-
vival distribution, Kaplan-Meier and Nelson-Aalen estimates, log-rank
test, Cox’s proportional hazard model. Parametric survival models for
Exponential, Gamma, Wiebull distributions. (12)
Bayesian Inference and Modeling: Prior and posterior distributions,
Bayesian models, Bayesian regression, Hierarchical Bayes models (10).
Introduction to mixture models (4)
Illustration of the methodology with real data.

Suggested References:
a) Applied Logistic Regression - Hosmer, D.W. and Lemeshow, S.
b) An Introduction to Categorical Data Analysis - Agresti, A.
c) Analysis of Categorical Data with R - Bilder, C. R. and Loughlin,
T. M.
d) Visualizing Categorical Data - Friendly, M.
e) Survival Analysis: A Self-Learning Text - Kleinbaum, D. G. and
Klein, M.
f) Applied Survival Analysis: Regression Modeling of time-to-event
data. - Hosmer, D.W., Lemeshow, S. and May, S.
g) Log Linear Models and Logistic Regression - Christensen, R.
h) Bayesian Statistical Modelling - Congdon, P.
i) Bayesian Data Analysis - Gelman, A., Carlin, J. et al.
j) Bayesian Essentials with R - Marin, J. and Robert, C.
k) Applied Bayesian Hierarchical Methods - Congdon, P.

18
9. Time Series and Statistical Finance

Time Series (28)

Exploratory analysis and graphical display; trend, seasonal and cyclical

components. Decomposition of time series into components, Smooth-
ing. (5)
Stationary Time Series: Brief Introduction to AR, MA and ARMA
models; Box-Jenkins correlogram analysis, ACF and PACF, introduc-
tion to periodogram, choice of AR and MA orders. (11)
Non-Stationary Time Series: introduction to ARIMA model; deter-
ministic and stochastic trends; introduction to ARCH and GARCH
models. (8)
Forecasting: basic tools, using exponential smoothing and Box-Jenkins
method. Residual analysis.(4)
Illustration of the methodology with real data.

Suggested References:
a) Introduction to Time Series and Forecasting - Brockwell, P. and
Davis R. A.
b) Analysis of Time Series - Chatfield, C.
c) Time Series Analysis and Its Applications with R - Shumway, R.H.
and Stoffer, D.S.
d) Introduction to Time Series Analysis and Forecasting - Mont-
gomery, D.C., Jennings, C.L., Kulachi, M.
e) Forecasting: Methods and Applications - Makridakis, S.G., Wheel-
wright, S.C. and Hyndman, R.J.

Statistical Finance (28)

Introduction to stock prices, returns and log-returns. Distribution of
returns, Assessing Normality using skewness, kurtosis and q-q plots. (4)
Market return and risk free rate. Capital Asset Pricing Model (CAPM).
Estimating beta and testing for CAPM. (6)
Options. Arbitrage and risk-neutral measure. European and American
options. Option pricing using Binomial model: 1 and 2 step. Black-
Scholes model (statement only), interpretation of drift and volatil-
ity. (8)
Value at risk and expected shortfall. Quantile estimation. Estimation
of tail-index. (6)

19
 Markowitz Portfolio Theory. Resampling for assessing estimation of
Efficient Portfolio. (4)
Illustration of methodology with real data.

Suggested References:
a) Statistics and Finance - Ruppert, David
b) Statistical Analysis of Financial Data in R - Carmona, Rene
c) Options, Futures and other Derivatives - Hull, John C. and Basu,
S.
d) Risk-Neutral Valuation - Bingham, N. and Keisel, R.

10. Project

————————————

20