49.

AGRICULTURAL STATISTICS
Unit 1: Probability and Mathematical Statistics
Elements of measure theory, Borel field, Probability measure; Random variable, Axiomatic
approach to probability; Laws of addition and multiplication; Bayes’ theorem. Discrete and
continuous variables; Mathematical expectation; Mathematical expectation of functions of
random variables; Moment generating function, Characteristic function; Raw and central
moments. Functions of random variables; Distribution function and its properties. Univariate
and bivariate probability distributions; Conditional and marginal distributions; Independence
of random variables; Transformation of random variables. Cauchy-Sehwarz inequality,
Jensen inequality, Markov inequality. Chebyshev’s inequality; Bernoulli weak law of large
numbers; Kolmogorov strong law of large numbers; Central limit theorem; Demoviere-
Laplace central limit theorem. Bernoulli, Binomial, Poisson, Negative binomial, Geometric,
Hypergeometric and Uniform distributions. Rectangular, Normal, Exponential, Gamma, Beta,
Cauchy and Lognormal distributions. Bivariate normal distribution. Probability distributions
of functions of random variables. Exponential Family of distributions. Mean and variance of
above mentioned distributions. Sampling distributions; Distribution of mean, difference
between two means and correlation coefficient; Central t, F and chi-square distributions, their
properties and interrelationships; Variance stabilizing transformations. Correlation and
Regression. Multiple and partial correlation coefficients; Order statistics; Distribution of rth
order statistic; Joint distribution of several order statistics and their functions; Distribution of
range and median.
Unit 2: Statistical Inference
Point estimation: Unbiasedness, Consistency, Sufficiency, Completeness; Neyman
factorization theorem with application; Minimum variance unbiased estimator; Cramer Rao
inequality; Rao Blackwell theorem. Methods of estimation: Method of moments, Method of
minimum chi-square, Method of maximum likelihood, their properties and applications.
Confidence interval estimation for parameters of Normal, Exponential, Binomial and Poisson
distributions. Testing of hypothesis; Neyman Pearson lemma; Unbiased test; Uniformly most
powerful unbiased tests and their constructions. One and two-sample tests about mean,
variance, proportion, simple correlation coefficient and simple regression coefficient;
Behrens-Fisher problem; Bartlett’s chi-square test; Likelihood ratio test and its asymptotic
properties. Chi-square tests of goodness of fit and independence. Non-parametric tests, One
and two sample sign and Wilcoxon sign rank tests, run test for randomness, Wilcoxon-Mann-
Whitney U test, Kruskal-Wallis and Friedman's tests, Kendall's coefficient of concordance.
Elements of sequential analysis; Wald's sequential probability ratio test.
Unit 3: Multivariate Analysis
Concept of random vector, Expectation operator, Dispersion matrix, Independence of random
vectors. Multinomial distribution. Multivariate normal distribution, Marginal, joint and
conditional distributions. Sample mean vector and its distribution; Maximum likelihood
estimates of mean vector and dispersion matrix. Tests of hypotheses about mean vector.
Wishart distribution and its properties; Hotelling's T2 and Mahalanobis’ D2 statistics; Null
distribution of Hotelling's T2. Multivariate analysis of variance; Wilk's lambda criterion and
its properties; Discriminant analysis, Computation of linear discriminant function (LDF),
Classification between two multivariate normal populations based upon LDF and
Mahalanobis’ D2. Canonical correlations; Factor analysis; Principal component analysis;
Cluster analysis, Similarities and Dissimilarities, Hierarchical clustering, Single and complete
linkage methods of clustering. Path analysis and computation of path coefficients; Multi-
dimensional scaling; Hierarchical and non- hierarchical clustering algorithm.

pg. 128
Unit 4: Design of Experiments
Theory of linear estimation; Gauss Markoff theorem; Aitkin's transformation; Hypothesis
testing and analysis of variance; Analysis of covariance; Random, fixed and mixed effects
models. Basic principles of design of experiments; Orthogonality; Contrast, Mutually
orthogonal contrasts. Completely randomized, Randomized complete block and Latin square
designs; Missing plot technique; Orthogonal and mutually orthogonal Latin squares; Graeco
Latin square designs. Balanced incomplete block (BIB) designs, General properties, Analysis
without and with recovery of intra-block information, Construction of BIB designs; Partially
balanced incomplete block (PBIB) designs with two associate classes, General properties;
Youden square designs; Cross-over designs. Factorial experiments, Confounding in 2n and 3n
factorial experiments, Partial and total confounding; Fractional factorial designs for
symmetrical factorials. Asymmetrical factorials. Split-plot and strip-plot designs. Combined
analysis of experiments. Designs for fitting first order and second order response surfaces,
Second order rotatable designs. Multiple comparison procedures; Sampling in field
experiments.
Unit 5: Sample Surveys
Complete survey vs sample survey; Probability sampling vs purposive sampling; Sampling
error; Sample space, Sampling design, Sampling strategy; Confidence interval. Simple
random sampling with and without replacement, Estimation of population mean and
population proportion; Inverse sampling; Stratified random sampling, Optimum allocation,
Number of strata, Construction of strata boundaries. Determination of sample size. Ratio,
regression and product methods of estimation; Separate and combined ratio estimators;
Cluster sampling; Multi-stage sampling with equal probability of selection of Units at each
stage; Two-phase sampling; Successive sampling over two occasions. Probability
proportional to size sampling - Cumulative method and Lahiri's method of selection; Horvitz
Thompson estimator, Ordered and unordered estimators, Sampling strategies due to
Midzuno-Sen and Rao-Hartley-Cochran; Inclusion probability proportional to size sampling.
Systematic sampling; Probability proportional to size systematic sampling. Non-sampling
errors, sources and classification, Non-response in surveys; Response error, Interpenetrating
sub-samples, Imputation methods; Warner's randomized response technique.
Unit 6: Statistical Genetics
Physical basis of inheritance, Segregation and Linkage; Analysis of segregation, Detection
and estimation of linkage for qualitative characters; Amount of information about linkage;
Combined estimation, Disturbed segregation. Gene and genotypic frequencies; Random
mating; Hardy-Weinberg law of equilibrium; Disequilibrium due to linkage for two pairs of
genes and sex-linked genes; Forces affecting gene frequency; Equilibrium between forces in
large populations, Polymorphism; Fisher’s fundamental theorem of natural selection;
Random genetic drift; Effect of finite population size. Polygenic system for quantitative
characters; Average effect of gene; Average effect of gene substitution; Dominance
deviation; Breeding value; Epistatic interaction deviation; Genotype- environment
correlation, genotype-environment interaction and its application; Multiple allelism in
continuous variations; Maternal effects; Different components of genetic variance and their
partitioning; Effect of inbreeding on quantitative characters; Heterosis; Inbreeding
depression; Effect of inbreeding on mean and variance of quantitative characters.
Resemblance between relatives; Phenotypic and genetic covariance between different
relatives; Concept and estimation of genetic parameters; Heritability, Repeatability and
Genetic correlation; Response due to selection, Selection index and its applications in plant
and animal genetic improvement programmes; Correlated response to selection; Restricted
selection index. Survival analysis.

pg. 129
Unit 7: Regression Analysis
Simple and multiple linear regression models and their analysis; Estimation and testing of
regression parameters, Sub-hypothesis testing, Restricted estimation; Polynomial regression:
Use of orthogonal polynomials. Use of dummy variables. Regression with ordinal data.
Selection of variables, Stepwise and Stagewise regressions. Regression diagnostics;
Adequacy and validation of models. Examination of residuals - specification error, auto-
correlation, Durbin-Watson statistic, Heteroscedasticity, Multi-collinearity. Weighted Least
Squares. Components of time-series. Fitting of different trend models. Autocorrelation and
Partial Auto-Correlation functions. Correlogram. Determination of cyclical variations.
Periodogram analysis. Linear Stationary models - Auto-Regressive, moving average and
mixed processes. Linear non-stationary models. Forecasting. Indirect Least Squares. Pooling
of cross-section and time-series data. Demand and Supply curves. Determination of demand
curves from market data. Engel’s curves. Pareto curves.
Unit 8: Mathematical Methods in Statistics and Optimization Techniques
Limit and continuity; Differentiation of functions, Successive differentiation, Partial
differentiation. Mean value theorems, Taylor and Maclaurin's series; Integration of rational,
irrational and trignometric functions. Differential equations of first order, Linear differential
equations of higher order with constant coefficients. Simple interpolation; Divided
differences; Numerical differentiation and integration. Group, Ring, Field and Vector spaces,
Subspaces, Basis, Galois field, Fermat's theorem and primitive elements. Linear
independence and dependence of vectors, Row and column spaces; Submatrices and
partitioned matrices; Determinant, rank and inverse of a matrix; Determinant and inverse of
partitioned matrices; Special matrices - Unitary, Similar, Hadamard, Circulant, Helmert's,
Idempotent and Orthogonal. Eigen values and eigen vectors. Spectral decomposition of
matrices. Kronecker and Hadamard product of matrices, Kronecker sum of matrices,
Permutation matrices, Full rank factorization. Generalized inverses, Moore-Penrose inverse,
Applications of generalized inverse; Generalized inverse of partitioned matrices; Solutions of
linear equations, Equations having many solutions; Spectral decomposition of matrices;
Quadratic forms. Optimization Techniques and Soft Computing: Classical optimization
techniques. Constrained optimization. Optimization and inequality. Numerical methods of
optimization. Direct Search method, Sequential Search method, Random Search method,
Simplex Search method, Gradient method and Method of Steepest Ascent. Linear
Programming Techniques - Simplex method, Duality and sensitivity analysis. Two-person
zero-sum game and linear programming; Integer Programming. Statistical applications. Non-
linear programming. Kuhn-Tucker conditions. Quadratic programming. Elements of Multiple
objective programming. Dynamic programming. Optimal control theory. Soft computing
tools - Artificial Neural Network, Support vector machines and probabilistic reasoning.
Genetic algorithm, decision tree, Bayes classifiers, Fuzzy Logic. Rough Set. Simulation
methods for various probability models. Resampling techniques: Jackknife and Bootstrap;
Monte Carlo simulation.

pg. 130