Hi Judy and others,
For assignment 3, question 1b, are we calculating standard errors and not standard deviations for the data (after means and SD for the raw data), given that the m=2, etc., variables are distributions of the sample means, and not the original distribution? Or does it not matter that we call is standard deviation or standard error?
If yes, the only way I see how to generate standard errors is through regress command. Is that correct?
Thanks, Rena
Ignore the second part of my question, of how to generate SE...
Rena, you should be calculating means and standard deviations for each sample for 1b (not standard errors).
I not sure I understand this.
For the raw data, I get why you would want the mean and standard deviation. However, for m=2, m=5, etc, we're looking at samples of means. In that case, wouldn't you want to get the standard error aka standard error of the mean?
Also, part (c) asks for "...what happens to the mean and standard deviation of the sample means?"
I'm interpreting this as asking for the standard error of the mean based on Lecture 4, slide 22 which states: "σ/√n is called the standard error, or more precisely, the standard error of the mean, and it is the standard deviation of the distribution of the sample mean".
I just want to make sure it is the standard deviations (σ) of BOTH raw data and samples you're asking for, since that is a different code than the one used for standard error. Thanks!
As Jenise said, you should be calculating means and standard deviations of each sample.
The standard deviation of sample means is called standard error of the mean, that is the standard error is a type of standard deviation.
Fatemah is correct. I want you to look at the standard deviation of the set of sample means you have. The standard deviation of a sample mean is also called the standard error of the mean, it is just a special name for the standard deviation of a distribution of sample means. You need to compare how these differ in magnitude as n increases.
Hi again (sorry for spamming your inboxes). I realized I probably confused everyone with my last email.
I think I may be understanding this question a bit more...
To clarify, the raw poisson data is a set of just numbers, and the m2, m10, etc are sets of the MEANS of these numbers.
So technically you want a command that gives you just the standard deviations, since for the raw data you want just the standard dev and for the sample means you want the standard deviations of the means (so in this case you will already have the means and will be needing the SD, not the SE).
I think the original confusion is not the difference between standard deviation or standard error, but what command to use when one gives you numbers labeled as standard deviation while another gives you numbers labeled as standard errors (intuitively you'd go with the second one, but not in this case).
Thanks!
