Scribd
Upload
32 views3 pages

AE 248: AI and Data Science: Prabhu Ramachandran 2024-03-01

The document discusses sampling statistics and the central limit theorem. It covers topics like sample mean, sample variance, sampling from a normal population, and the distribution of the sample mean and variance. The central limit theorem states that the sum of iid random variables will be approximately normally distributed for large n, even if the individual variables are not normally distributed. The document provides equations and explanations for concepts like expected value and variance of the sample mean and variance.

Uploaded by

prasan0311das
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
32 views3 pages

AE 248: AI and Data Science: Prabhu Ramachandran 2024-03-01

The document discusses sampling statistics and the central limit theorem. It covers topics like sample mean, sample variance, sampling from a normal population, and the distribution of the sample mean and variance. The central limit theorem states that the sum of iid random variables will be approximately normally distributed for large n, even if the individual variables are not normally distributed. The document provides equations and explanations for concepts like expected value and variance of the sample mean and variance.

Uploaded by

prasan0311das
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

AE 248: AI and Data science

Sampling Statistics

Prabhu Ramachandran

2024-03-01

Sample

• 𝑋1 , 𝑋2 , ...𝑋𝑛 are independent


• with common distribution 𝐹
• Parametric inference: assume distribution 𝐹 but without specifics
• Non-parametric inference

Sample mean

• Consider a population with given mean and variance


• Take a sample of 𝑛 elements

𝑋 + ... + 𝑋𝑛
𝑋̄ = 1
𝑛
. . .
̄
• What is 𝐸[𝑋]?
̄
• What is Var(𝑋)?

The Central Limit Theorem

• If 𝑋1 , 𝑋2 , ...𝑋𝑛 are iid each with 𝜇 and 𝜎


• then for 𝑛 large,
• ∑𝑖 𝑋𝑖 is approximately normal with mean 𝑛𝜇 and variance 𝑛𝜎2

1
The Central Limit Theorem

• Recall previous chapter and binomial distributions


• Even a binomial distribution may be approximated to be normal
• So how large should 𝑛 be?

Sample Variance
𝑛
2
∑𝑖=1 (𝑋𝑖 − 𝑋)̄ 2
𝑆 =
𝑛−1
• Compute 𝐸[𝑆 2 ]

Sample Variance

Using:
𝑛 𝑛
• ∑𝑖=1 (𝑥𝑖 − 𝑥)̄ 2 = ∑𝑖=1 𝑥2𝑖 − 𝑛𝑥2̄
• 𝐸[𝑋 2 ] = 𝑉 𝑎𝑟(𝑋) + 𝐸[𝑋]2
• Show that: 𝐸[𝑆 2 ] = 𝜎2

Sampling from a Normal Population

• Population is normal with 𝜇, 𝜎


• How are 𝑋,̄ 𝑆 2 distributed?

Distribution of Sample Mean

• 𝑋̄ is normally distributed
• 𝐸[𝑋]̄ = 𝜇
• 𝑉 𝑎𝑟[𝑋]̄ = 𝜎2 /𝑛

Joint Distribution of mean and variance

• 𝑋̄ and 𝑆 2 are independent


• (𝑛 − 1)𝑆 2 /𝜎2 is 𝜒2𝑛−1

𝑛 𝑛
∑(𝑥𝑖 − 𝑥)̄ 2 = ∑(𝑥𝑖 − 𝜇)2 − 𝑛(𝑥̄ − 𝜇)2
𝑖=1 𝑖=1

2
Distribution of 𝑆 2
𝑛 √ 2
𝑛
𝑋𝑖 − 𝜇
2
∑𝑖=1 (𝑋𝑖 − 𝑋)̄ 2 𝑛(𝑋̄ − 𝜇)
∑( ) = −[ ]
𝑖=1
𝜎 𝜎2 𝜎
. . .

• LHS: 𝜒2𝑛
• RHS, second term, 𝜒21

Corollary

• Sample of 𝑛 items from a normal population

√ (𝑋̄ − 𝜇)
𝑛 ∼ 𝑡𝑛−1
𝑆
. . .
√
• Recall: 𝑛𝑍/√𝜒2𝑛

You might also like