This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Administration I have regraded midterms. Come get them after class. Joke. Joke. Why do computer scientists confuse Halloween and Christmas? Joke. Why do computer scientists confuse Halloween and Christmas? OCT 31 = DEC 25 . Joke. Why do computer scientists confuse Halloween and Christmas? OCT 31 = DEC 25 . , CS70: Satish Rao: Lecture 28. 1. Polling. 2. Law of Large Numbers. 3. Joint Distributions Tail Bounds. Markov: For a nonnegative random variable X , Pr [ X ≥ c ] ≤ E ( X ) c . Tail Bounds. Markov: For a nonnegative random variable X , Pr [ X ≥ c ] ≤ E ( X ) c . Chebyshev’s Inequality: For a random variable X with expectation μ Pr [  X μ  ≥ α ] ≤ Var ( X ) α 2 . Tail bounds!!! Pr [ X ≥ i ] Chebyshev bound. Markov bound. . 5 1 . 15 30 45 60 75 90 105 120 135 150 Tail bounds!!! Pr [  X μ  ≥ i ] Chebyshev bound. Markov bound. . 5 1 . 15 30 45 60 75 90 105 120 135 150 Polling. Proportion of democrats? Polling. Proportion of democrats? Polling. Proportion of democrats? Poll: Choose a n people at random and average. Polling. Proportion of democrats? Poll: Choose a n people at random and average. How many people to get proportion within . 1 Polling. Proportion of democrats? Poll: Choose a n people at random and average. How many people to get proportion within . 1 with probability at least . 95? The setup. S nnumber who are dems in sample. The setup. S nnumber who are dems in sample. A n = 1 n S n . The setup. S nnumber who are dems in sample. A n = 1 n S n . X i = 1 if i th person is democrat otherwise The setup. S nnumber who are dems in sample. A n = 1 n S n . X i = 1 if i th person is democrat otherwise S n = ( X 1 + X 2 + ··· + X n ) . The setup. S nnumber who are dems in sample. A n = 1 n S n . X i = 1 if i th person is democrat otherwise S n = ( X 1 + X 2 + ··· + X n ) . A n = 1 n S n . The setup. S nnumber who are dems in sample. A n = 1 n S n . X i = 1 if i th person is democrat otherwise S n = ( X 1 + X 2 + ··· + X n ) . A n = 1 n S n . E ( A n ) = 1 n E ( S n ) The setup. S nnumber who are dems in sample. A n = 1 n S n . X i = 1 if i th person is democrat otherwise S n = ( X 1 + X 2 + ··· + X n ) . A n = 1 n S n . E ( A n ) = 1 n E ( S n ) = 1 n ( pn ) = p . The setup. S nnumber who are dems in sample. A n = 1 n S n . X i = 1 if i th person is democrat otherwise S n = ( X 1 + X 2 + ··· + X n ) . A n = 1 n S n . E ( A n ) = 1 n E ( S n ) = 1 n ( pn ) = p . The fraction of democrats! The setup. S nnumber who are dems in sample. A n = 1 n S n . X i = 1 if i th person is democrat otherwise S n = ( X 1 + X 2 + ··· + X n ) . A n = 1 n S n . E ( A n ) = 1 n E ( S n ) = 1 n ( pn ) = p . The fraction of democrats! Unbiased estimator!!! The setup....
View
Full
Document
This note was uploaded on 02/29/2012 for the course COMPSCI 70 taught by Professor Rau during the Fall '11 term at Berkeley.
 Fall '11
 Rau

Click to edit the document details