Zembower, Kevin
2007-Oct-16 20:30 UTC
[R] Calculating confidence in an estimate including number of trials?
[Yes, this is related to a homework problem, but is not the problems itself.] In my mathematical statistics class, we've just learned about properties of estimators, and I can now solve manually problems like this: A sample of size n = 16 is drawn from a normal distribution where sigma = 10 but mu is unknown. If mu = 20, what is the probability that the estimator mu hat = Y bar will lie between 10.0 and 21.0?[1] I solved this by converting to Z scores and using a table of cumulative values under the normal curve and got an answer of .3108 (someone please tell me if I'm wrong). Now I'd like to know how to use R to solve this type of problem. In all my other problems using normal curves, I used dnorm or pnorm, but neither of these includes anything regarding the number of trials. I can put the math into R after I've worked out the equation, but I wondered if there was an R function that computed this directly, in the same fashion that pnorm can compute probabilities using parameters of mean and sd. Using help.search for 'estimator' or 'sample mean' didn't turn up anything that I recognized. Any hints on where to go looking for this? Thanks for your help and advice. -Kevin Kevin Zembower Internet Services Group manager Center for Communication Programs Bloomberg School of Public Health Johns Hopkins University 111 Market Place, Suite 310 Baltimore, Maryland 21202 410-659-6139 [1] Introduction to Mathematical Statistics and its applications, Larsen and Marx, fourth ed., question 5.4.4.
Daniel Lakeland
2007-Oct-16 20:35 UTC
[R] Calculating confidence in an estimate including number of trials?
On Tue, Oct 16, 2007 at 04:30:48PM -0400, Zembower, Kevin wrote:> Now I'd like to know how to use R to solve this type of problem. In all > my other problems using normal curves, I used dnorm or pnorm, but > neither of these includes anything regarding the number of trials.pnorm can be used like your table of area under the normal curve. To account for size of sample you have to scale the variance appropriately according to the theory you have learned in your course. -- Daniel Lakeland dlakelan at street-artists.org http://www.street-artists.org/~dlakelan