A list of likely exam questions.
- Discuss the difference between the definition of probability
in probability space and empirical probability.
- "Prove" Bayes' theorem and discuss its relevance in inference
problems.
- Describe the concept of convergence in probability theory.
- Prove Chebyshev's inequality and use it to find the weak law
of large numbers.
- Describe the link between probability generating functions and
momenta of discrete distributions.
- Find the probability generating function of a sum of random
variables.
- Find the probability generating function of common
distributions, Poisson, Bernoulli, Binomial.
- Find the characteristic function of the Gaussian distribution.
- Give an outline of the proof of the Central Limit Theorem.
- Describe the geometric distribution.
- Describe the negative binomial distribution.
- Describe the hypergeometric distribution.
- Describe the elementary model of cancer onset vs. age.
- Find the formula for the detected rate vs. true rate for
paralizable and non-paralizable counters.
- Prove the formula for the coincidence rate of multiple
detectors.
- Find the high-rate limit of the Poisson distribution.
- Describe in detail the multivariate Gaussian distribution.
- Describe the gamma distribution and prove the formulas for
expectation value and variance.
- Demonstrate the formula for the distribution of distance to
the nearest neighbor (in the limit of large number density).
- Repeat the proof that the correlation coefficient is lies in
the (-1,+1) interval.
- Find the probability distribution of the sample mean.
- Explain order statistics.
- What is a box plot?
- What is a kernel density estimate?
- What is an empirical probability density?
- Discuss the algorithmic generation of pseudo-random numbers.
- Describe the transformation method to generate arbitrarily
distributed pseudo-random numbers
- Describe the acceptance-rejection method and provide an
elementary proof of its validity.
- Describe the statistical bootstrap.
- Describe the principle of maximum likelihood (MaxL).
- Discuss the application of MaxL to normally distributed data,
to estimate mean and/or standard deviation.
- Discuss the application of MaxL to data extracted from a
Poisson distribution to estimate the mean.
- Prove the Bartlett identities.
- The Cramer-Rao-Fisher bound and the Fisher information.
- Define the Shannon information.
- How to use the likelihood to estimate the confidence intervals
of model parameters.
- Describe the method of least squares.
- Describe linear regression.
- How to use a statistic to discriminate between null and
alternative hypothesis.
- Prove the Neyman-Pearson lemma.
- What are p-values and how do you use them in multiple tests of
the null hypothesis?
- Describe Wilks' theorem.
- Describe the Naive Bayesian Classification.
- Describe the k-means algorithm.
- Describe the principal component method.
December 2023.