Data Science Past Paper

• Foundations of Data Science (2020–2018)

Probability Models

• Mock Exam1
• y2022p6q7 (a.i.ii)
• y2022p6q8 (c,d)
• y2021p6q7
  • estimate parameters
• y2021p6q8
• y2019p6q8
  • linear model
• y2018p6q8

Inference

• Mock Exam3
• y2020p6q8
  • Inference for significance
    • parametric resampling
    • multiple unknown parameters' MLE
      • use the entire dataset (Multinomial)
    • one-sided vs two-sided test, p-value
  • linear vs count-based model
• y2022p6q7 (a.iii-c)
  • (non)parametric resampling, confidence interval
  • computational Bayesian methods

Bayesian

• Mock Exam2
• y2019p6q7

Stochastic processes

MC

• Mock Exam4
• y2020p6q7
  • Matrix $P, \pi$. Detailed balance equations.
  • posterior $\omega$, pseudocode
  • posterior confidence interval $prior_{\bar{sample}}(F_\omega<0.025,F_\omega>0.975)$
• y2022p6q8 (a,b)
  • stationary distribution, memoryless-ness