Date |
Content of the Lecture |
Assignments/Readings/Notes |
30/07 (Tue) |
- Introduction, course overview and course policies
Descriptive Statistics
- Descriptive statistics: key terminology
- Methods to represent data: frequency tables, bar/line graphs, frequency polygon, pie-chart
- Concept of frequency and relative frequency
- Cumulative frequency plots
- Interesting examples of histograms of intensity values in an image
- Data summarization: mean and median
- "Proof" that median minimizes the sum of absolute deviations - using calculus
|
|
02/08 (Fri) |
- Proofs that median minimizes the sum of absolute deviations without using calculus
- Concept of quantile/percentile
- Standard deviation and variance, some applications
- Two-sided Chebyshev inequality with proof; One-side Chebyshev inequality (Chebyshev-Cantelli inequality)
- Concept of correlation coefficient and formula for it; proof that its value lies from -1 to +1
|
|
03/08 (Sat) |
- Correlation coefficient: properties; uncentered correlation coefficient; limitations of correlation coefficient and Anscombe's quartet
- Correlation and causation
- Proof of one-sided Chebyshev's inequality
|
|
06/08 (Tue) |
Discrete Probability
- Discrete probability: sample space, event, composition of events: union, intersection, complement, exclusive or, De Morgan's laws
- Boole's and Bonferroni's inequalities
- Conditional probability, Bayes rule, False Positive Paradox
- Birthday paradox
- Independent and mutually exclusive events
|
|
09/08 (Fri) |
Random variables
- Random variable: concept, discrete and continuous random variables
- Probability mass function (pmf), cumulative distribution function (cdf) and probability density function (pdf)
- Expected value for discrete and continuous random variables
- Law of the Unconscious Statistician (LOTUS): Expected value of a function of a random variable
- The mean and the median as minimizers of squared and absolute losses respectively (with proof for the former)
- Variance and standard deviation, with alternate expressions
|
|
13/08 (Tue) |
- Concept of joint PMF, PDF, CDF
- Concept of covariance, concept of mutual independence and pairwise independence
- Properties of covariance
- Covariance: properties, correlation versus independence
- Concept of moment generating function, two different proofs of uniqueness of moment generating function for discrete random variables, properties of moment generating functions
- PDF/PMF of sum of random variables
- Proof that median minimizes total absolute loss
|
|
16/08 (Fri) |
- Conditional CDF, PMF, PDF; verification of definition
Families of random variables
- Concept of families of random variables
- Bernoulli PMF: mean, median, mode, variance, MGF
- Binomial PMF: relation to Bernoulli PMF, mean, median, mode, variance, plots, MGF, difference between binomial and geometric distribution
- Gaussian (normal) PDF: motivation from the central limit theorem
- Illustration of central limit theorem, statement of central limit theorem
|
|
17/08 (Sat) |
- Derivation of mean, variance, MGF, median, mode
- CDF of a Gaussian and its relations to error functions; probability of a Gaussian random variable to have values between mu +/- k sigma.
- Statement of central limit theorem and its extensions; proof of CLT using MGF
- application of CLT and its relation to the binomial distribution - de Moivre-Laplace theorem (without proof);
- one application of the CLT; relation between CLT and the law of large numbers
|
|
20/08 (Tue) |
- Gaussian tail bounds
- Distribution of the sample mean and the sample variance, Bessel's correction;
- Chi-squared distribution - definition, genesis, MGF, properties; use of a chi-square distribution toward defining the PDF of the sample variance
- Uniform distribution: mean, variance, median, MGF; applications in sampling from a pre-specified PMF; application in generating a random permutation of a given set
|
|
23/08 (Fri) |
- Poisson distribution: mean, variance, MGF, mode, addition of Poisson random variables, examples; derivation of Poisson from binomial
- Relation between Poisson and Gaussian distributions, examples
- Multinomial PMF - generalization of the binomial, mean vector and covariance matrix for a multinomial random variable, MGF for multinomial
|
|
24/08 (Sat) |
- Exponential distribution: mean, median, MGF, variance, property of memorylessness, minimum of exponential random variables
Parameter Estimation
- Parameter Estimation
- Concept of parameter estimation (or parametric PDF/PMF estimation)
- Maximum likelihood estimation (MLE)
- MLE for parameters of Bernoulli, Poisson, Gaussian and uniform distributions
- Least squares line fitting as an MLE problem
|
|
27/08 (Tue) |
- Least squares line fitting as an MLE problem
- Concept of estimator bias, mean squared error, variance
- Estimators for interval of uniform distribution: example of bias
- Concept of two-sided confidence interval and one-sided confidence interval
- Confidence interval for mean of a Gaussian with known standard deviation
- Confidence interval for variance of a Gaussian
|
|
30/08 (Fri) |
- Concept of nonparametric density estimation
- Concept of histogram as a probability density estimator
- Bias, variance and MSE for a histogram estimator for a smooth density (with bounded first derivatives) which is non-zero on a finite-sized interval; derivation
of optimal number of bins (equivalently, optimal binwidth) and optimal MSE O(n^{-2/3})
|
|
13/09 (Fri) |
- Hypergeometric distribution: genesis, mean, variance
- Applications of the hypergeometric distribution in counting of animals via the capture-recapture method
- Concept of kernel density estimator
- Bias, variance and MSE for a kernel density estimator for a smooth density (with bounded second derivatives); derivation
of optimal bandwidth and optimal MSE O(n^{-4/5})
|
|