inference - Advanced Quantitative Research Methodology,...

Info iconThis preview shows pages 1–15. Sign up to view the full content.

View Full Document Right Arrow Icon
Advanced Quantitative Research Methodology, Lecture Notes: Theories of Inference 1 Gary King http://GKing.Harvard.Edu January 31, 2010 1 c ± Copyright 2010 Gary King, All Rights Reserved. Gary King () January 31, 2010 1 / 1
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
The Problem of Inference Gary King () Inference 2 / 1
Background image of page 2
The Problem of Inference 1. Probability: P( y | M ) = P( known | unknown) Gary King () Inference 2 / 1
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
The Problem of Inference 1. Probability: P( y | M ) = P( known | unknown) 2. The goal of inverse probability: P( M | y ) = P(unknown | known ). Gary King () Inference 2 / 1
Background image of page 4
The Problem of Inference 1. Probability: P( y | M ) = P( known | unknown) 2. The goal of inverse probability: P( M | y ) = P(unknown | known ). 3. Define: M = { M * , θ } , where M * is assumed and θ is to be estimated Gary King () Inference 2 / 1
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
The Problem of Inference 1. Probability: P( y | M ) = P( known | unknown) 2. The goal of inverse probability: P( M | y ) = P(unknown | known ). 3. Define: M = { M * , θ } , where M * is assumed and θ is to be estimated 4. More limited goal: P( θ | y , M * ) P( θ | y ). Gary King () Inference 2 / 1
Background image of page 6
The Problem of Inference 5. Bayes Theorem (no additional assumptions): Gary King () Inference 3 / 1
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
The Problem of Inference 5. Bayes Theorem (no additional assumptions): P( θ | y ) = P( θ, y ) P( y ) [Defn. of conditional probability] Gary King () Inference 3 / 1
Background image of page 8
The Problem of Inference 5. Bayes Theorem (no additional assumptions): P( θ | y ) = P( θ, y ) P( y ) [Defn. of conditional probability] = P( θ )P( y | θ ) P( y ) [P( AB ) = P( B )P( A | B )] Gary King () Inference 3 / 1
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
The Problem of Inference 5. Bayes Theorem (no additional assumptions): P( θ | y ) = P( θ, y ) P( y ) [Defn. of conditional probability] = P( θ )P( y | θ ) P( y ) [P( AB ) = P( B )P( A | B )] = P( θ )P( y | θ ) R P( θ )P( y | θ ) d θ [P( A ) = Z P( AB ) dB ] Gary King () Inference 3 / 1
Background image of page 10
The Problem of Inference 5. Bayes Theorem (no additional assumptions): P( θ | y ) = P( θ, y ) P( y ) [Defn. of conditional probability] = P( θ )P( y | θ ) P( y ) [P( AB ) = P( B )P( A | B )] = P( θ )P( y | θ ) R P( θ )P( y | θ ) d θ [P( A ) = Z P( AB ) dB ] 6. If we knew the right side, we could compute the inverse probability. Gary King () Inference 3 / 1
Background image of page 11

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
The Problem of Inference 5. Bayes Theorem (no additional assumptions): P( θ | y ) = P( θ, y ) P( y ) [Defn. of conditional probability] = P( θ )P( y | θ ) P( y ) [P( AB ) = P( B )P( A | B )] = P( θ )P( y | θ ) R P( θ )P( y | θ ) d θ [P( A ) = Z P( AB ) dB ] 6. If we knew the right side, we could compute the inverse probability. 7. 2 theories of inference arose to interpret this result: likelihood and Bayesian Gary King () Inference 3 / 1
Background image of page 12
The Problem of Inference 5. Bayes Theorem (no additional assumptions): P( θ | y ) = P( θ, y ) P( y ) [Defn. of conditional probability] = P( θ )P( y | θ ) P( y ) [P( AB ) = P( B )P( A | B )] = P( θ )P( y | θ ) R P( θ )P( y | θ ) d θ [P( A ) = Z P( AB ) dB ] 6. If we knew the right side, we could compute the inverse probability. 7. 2 theories of inference arose to interpret this result: likelihood and Bayesian 8. In both, P( y | θ ) is a traditional probability density ; what’s the rest? Gary King () Inference 3 / 1
Background image of page 13

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
The Problem of Inference 5. Bayes Theorem (no additional assumptions): P( θ | y ) = P( θ, y ) P( y ) [Defn. of conditional probability] = P( θ )P( y | θ ) P( y ) [P( AB ) = P( B )P( A | B )] = P( θ )P( y | θ ) R P( θ )P( y | θ ) d θ [P( A ) = Z P( AB ) dB ] 6. If we knew the right side, we could compute the inverse probability.
Background image of page 14
Image of page 15
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 312

inference - Advanced Quantitative Research Methodology,...

This preview shows document pages 1 - 15. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online