- Lecture template I Sample spaces, multinomial, repeated sampling (updated April 6)
- Lecture template II ML estimation for discrete distributions
- Lecture (and template) III Estimating small discrete probabilities (14 slides) -- small edits possible
Estimating the unseen: a
sublinear-sample canonical estimator of distributions Gregory Valiant and Paul Valiant
Frequencies of Chinese Characters a great example of discrete distribution with many small probabilities
The Wikipedia page of meteorite fall statistics by type, a distribution with small m but also small probabilities
- Lecture template IV Parametric ML estimation for continuous distributions
- Lecture template V Non-parametric estimation for continuous distributions
- Lecture VI Bayesian estimation
- Lecture VII Linear and logistic regression
- Lecture 8 Linear and logistic regression lecture notes
Old handouts, FYI
- Lecture 2 Bayesian inference with D
- Clustering (116 slides!!) what you need are the following
- Introduction (pages 1--7)
- K-means (pages 15,16, 20,21,22 and 11)
- Mixtures and the EM algorithm (pages 24-30, 32, 34)
- Selecting K (pages 44-47, 51-52)
- Non-parametric (pages 55-58, and 77)
Links to figures
|