How Do You Calculate the Probability of a Sample Mean in Normal Distribution?

Click For Summary

Discussion Overview

The discussion revolves around calculating probabilities related to sample means in normal and Poisson distributions. Participants explore the implications of assumptions such as independence and the applicability of the Central Limit Theorem in different contexts, including both theoretical and practical aspects of probability calculations.

Discussion Character

  • Exploratory
  • Technical explanation
  • Debate/contested
  • Mathematical reasoning

Main Points Raised

  • One participant presents a scenario involving normally distributed scores and attempts to calculate the probability of a sample mean being less than a certain value, but encounters difficulties with the z-score.
  • Another participant raises concerns about the assumption of independence between scores on different tests, suggesting that high scores on one may correlate with high scores on another.
  • A participant questions the appropriateness of approximating a Poisson distribution with a normal distribution, arguing that the mean is too small for a good approximation.
  • There is a discussion about the nature of Poisson outputs being discrete while the mean can be continuous, leading to confusion about calculating probabilities for continuous values.
  • Some participants express uncertainty about how to apply z-scores in the context of Poisson distributions, questioning their relevance and seeking clarification on the cumulative distribution function (CDF) for Poisson distributions.
  • One participant defends the use of the normal approximation based on the Central Limit Theorem, despite others suggesting that it may not be suitable for small means.
  • There is a request for clarification on how to compute probabilities for a sample mean in relation to the Poisson distribution, indicating a lack of consensus on the methodology.

Areas of Agreement / Disagreement

Participants express differing views on the appropriateness of using normal approximations for Poisson distributions and the implications of independence in the context of the problem. The discussion remains unresolved regarding the best approach to calculate the probabilities in question.

Contextual Notes

Participants highlight limitations related to assumptions of independence, the applicability of the Central Limit Theorem, and the challenges of using z-scores in non-Gaussian contexts. There are also unresolved questions about the correct application of the CDF for Poisson distributions.

Unusualskill
Messages
35
Reaction score
1
The scores X1 and X2 in papers 1 and 2 of an examination are normally distributed with means 24.3 and 31.2 respectively and standard deviations 3.5 and 3.1 respectively The final mark for each candidate is found by calculating 2X1+1.5X2. Find the probability that a random sample of 8candiates will have a mean final mark of less than 60.

what i have done so far is F~N(95.4,70.6225) and Sample mean of F, N (95.4,70.6225/8)
P(sample mean of F<60)
... then unable to do alrdy because the z-score is too large
 
Physics news on Phys.org
It seems to me that independence would be violated rather badly. :-) Students scoring highly on one test are more likely to score highly on the other. You're probably supposed to assume independence, but you should scold your teacher for not being explicit about it.
 
chogg said:
It seems to me that independence would be violated rather badly. :-) Students scoring highly on one test are more likely to score highly on the other. You're probably supposed to assume independence, but you should scold your teacher for not being explicit about it.

Chogg, can i ask you another question, which my answer is different from answer given.
The random variable X is distributed Po(4.1) .Find the probability that the mean of a random sample of 50 observations of X is less than 4.5.

first, i find out p=4.1/50 variance=npq=3.7638
then approx to normal, N(4.1,3.7638/50)
P(sample mean<4.5)
continuity correction with -1/100
anything wrong?what is the ans u get?
 
I assume "Po" means Poisson, and 4.1 is the value of \lambda.

Why would you switch to the approximating normal? It's going to be a rather lousy approximation with such a small mean. Why not just use the poisson directly?

Anyway, you might find it useful to know that the sum of (independent!) draws from Poisson distributions is Poisson-distributed, and its mean is the sum of the means. You also might find it useful to translate the statement about the mean to a statement about the sum of your 50 random draws.
 
But how to do Possion with continuous(4.5) ? r u able to get z-score for this ques 1.34?if yes, can you show me the step pls thanks a lot...if like u said using sum, then Po~(205) still need to use normal is it?
 
Last edited:
Unusualskill said:
But how to do Possion with continuous(4.5) ?

http://en.wikipedia.org/wiki/Poisson_distribution

The output of a Poisson is discrete: 0, 1, 2, etc.

The mean of a Poisson is a continuous quantity. It could be anything.

For example, imagine a probability distribution which gives 1 with probability p, and 0 with probability 1-p. It only ever outputs discrete numbers (1 or 0), but its mean is a continuous quantity (p).

Unusualskill said:
r u able to get z-score for this ques 1.34?if yes, can you show me the step pls thanks a lot...if like u said using sum, then Po~(205) still need to use normal is it?

Why do you care about z-scores? This is a Poisson problem, not a Gaussian problem. z-scores are for Gaussian distributions.

Also: what does "ques 1.34" mean? Is this a homework problem? (It sure sounds like it.) If it is, you're asking in the wrong forum.

In any case: if you want to know the probability that a random variable is less than some value, the concept that will help you is the cumulative distribution function, or "CDF". What is the CDF for the Poisson distribution? Does it give a reasonable-sounding answer?
 
chogg said:
http://en.wikipedia.org/wiki/Poisson_distribution

The output of a Poisson is discrete: 0, 1, 2, etc.

The mean of a Poisson is a continuous quantity. It could be anything.

For example, imagine a probability distribution which gives 1 with probability p, and 0 with probability 1-p. It only ever outputs discrete numbers (1 or 0), but its mean is a continuous quantity (p).



Why do you care about z-scores? This is a Poisson problem, not a Gaussian problem. z-scores are for Gaussian distributions.

Also: what does "ques 1.34" mean? Is this a homework problem? (It sure sounds like it.) If it is, you're asking in the wrong forum.

In any case: if you want to know the probability that a random variable is less than some value, the concept that will help you is the cumulative distribution function, or "CDF". What is the CDF for the Poisson distribution? Does it give a reasonable-sounding answer?

im using normal because of central limit theorem, which states that if n is large, the distribution can be approx to normal. Because i duno homework to do poisson for P(X<4.5)...
 
Unusualskill said:
im using normal because of central limit theorem, which states that if n is large, the distribution can be approx to normal. Because i duno homework to do poisson for P(X<4.5)...

Did you find the CDF for the Poisson distribution from the wikipedia page I linked to?

Did you try plugging in your numbers to that formula?
 
chogg said:
Did you find the CDF for the Poisson distribution from the wikipedia page I linked to?

Did you try plugging in your numbers to that formula?

i know that formula for poisson . P(x<4.5)=P(x=0,1,2,3,4)? but this is a question for sampling? or can you show me how you did?
 

Similar threads

  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
1
Views
4K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 9 ·
Replies
9
Views
5K
  • · Replies 6 ·
Replies
6
Views
4K
Replies
2
Views
2K
  • · Replies 6 ·
Replies
6
Views
4K
  • · Replies 9 ·
Replies
9
Views
2K