Stats question, how to get expected values

  • Thread starter Thread starter wtmoore
  • Start date Start date
  • Tags Tags
    Stats
Click For Summary
SUMMARY

The discussion focuses on estimating expected values for a random variable X that follows a normal distribution, specifically X~N(8,2). The key formulas discussed include the unbiased estimator for the mean, E(xbar) = (1/N)Σx_n, which confirms that E[xbar] equals the population mean μ. Additionally, the expected value for the sample variance is addressed with E(s^2) = E[{Σ(x_i - xbar)^2}/(n-1)], emphasizing the need to understand how summation interacts with expected values. Participants are encouraged to derive unbiased estimators for other parameters of the random variable.

PREREQUISITES
  • Understanding of normal distribution, specifically X~N(8,2)
  • Knowledge of expected value and unbiased estimators
  • Familiarity with sample mean and sample variance calculations
  • Basic statistics concepts, including summation notation
NEXT STEPS
  • Research the derivation of unbiased estimators for population parameters
  • Study the properties of expected values in statistics
  • Learn about the Central Limit Theorem and its implications for sample means
  • Explore advanced statistical concepts such as maximum likelihood estimation
USEFUL FOR

Students in statistics, data analysts, and anyone involved in statistical estimation and hypothesis testing will benefit from this discussion.

wtmoore
Messages
20
Reaction score
0

Homework Statement



I have taken a random sample such that X~N(8,2).

I want to use the samples that I have generated to estimate E(xbar), E(s2), E(α21) and E(α22) for the population.

Homework Equations


The Attempt at a Solution



I am not entirely sure how to do these.

I know that for a random variable, E(x)= (x1 + x2 ... +xn)/n, but I am not sure if this is the same for a sample/xbar.

I have googled and can't find the formulas, does anyone know them here?

I know for s^2 that E(s^2) = E[{sum(i to n) (xi-xbar)^2}/(n-1)]

so now I can bring the 1/(n-1) out front of the E, but I'm not sure how the summation acts.
 
Last edited:
Physics news on Phys.org
You are confusing some concepts. You want to estimate say, the mean of the random variable X μ=E{X}, you use the estimator [itex]\bar{x}=\frac{1}{N}\sum_nx_n[/itex] where x_n are the samples of X. You can prove [itex]E[\bar{x}]=\mu[/itex] so that the estimator which is a function of the samples has an expected value that is equal to the mean μ of X, so that this estimator is unbiased. Now try to find unbiased estimators of other parameters of the random variable X that you want to estimate, as a function of the samples, i.e., the expected values are equal to the parameters you want to estimate.
 

Similar threads

Replies
4
Views
2K
Replies
10
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
Replies
5
Views
2K
  • · Replies 3 ·
Replies
3
Views
3K
Replies
5
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K