Lecture07 - 2 May 2003 Biostatistics 6650--L7 1 Todays...

Info iconThis preview shows pages 1–10. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 2 May 2003 Biostatistics 6650--L7 1 Todays Schedule Estimation of the population mean Central limit theorem Interval estimation Mean of continuous data known unknown( t-distn) Binomial Proportion Poisson Sample size for CIs Variance( chi-square distn) HO: Geigy Bin CI/ Rosner T8 2 May 2003 Biostatistics 6650--L7 2 Estimation of Goal: Make a statement about the population mean Sample statistics , Median, mode, trimmed mean, range, s 2 , percentiles Natural estimator is , is a random variable. It can be shown that the E( ) from a sample of size n will equal . Thats good on average, but it doesnt say anything about a given sample. Can we describe a region where we think is based on our sample? X X X X 2 May 2003 Biostatistics 6650--L7 3 Estimation of If 2 is the variance of an individual X, what is the variance of ? n i i=1 n 2 i i=1 n 2 i i=1 2 2 2 Var{X}=Var{ X /n} =(1/n )Var{ X } =(1/n ) Var(X ), independent =(1/n )n = /n X 2 May 2003 Biostatistics 6650--L7 4 Estimation of Var( ) = 2 /n , Logical that variability in sample means should decrease with the sample size Standard Deviation of or Standard Error of the Mean(SEM) or Sampling distribution of X ~ N( , 2 ) ~N( , 2 /n ) Standardize : X X X Z n - = / , estimate with / SE n s n = X X X 2 May 2003 Biostatistics 6650--L7 5 Central Limit Theorem Central Limit Theorem If X is from an unknown distribution with mean and variance 2 , then for n large the sample mean will have an approximate normal distribution with mean and variance 2 /n. X ~ ?( , 2 ) N( , 2 /n) Extremely important for statistical inference. Often dont know the distributional form, but suspect non-normality Applies also to the sample sum in most cases n>25 is adequate for approximations to two decimal places-- Lindgren, Stat Theory, 3rd Ed X : & 2 May 2003 Biostatistics 6650--L7 6 CLT: Distn of Sample Means www.rci.rutgers/~cabrera/211/clt.gif n=1 n=5 n=50 n=500 2 May 2003 Biostatistics 6650--L7 7 Central Limit Theorem Think about X with a Binomial distribution X is the sum of n independent 0/1 variables the CLT applies to sums Normal approx to Binomial(and Poisson) is a special case of CLT 2 May 2003 Biostatistics 6650--L7 8 Interval Estimation: Continuous X When is known , an approximate 100*(1- )% confidence interval for the population mean is given by: Or by: Where, Z 1- /2 is 100*(1- )th percentile of the standard normal distribution 1 1 2 2 ME to ME, ME=margin of error [ / ] to [ / ] X X X Z n X Z n --- +- + 1 2 / X Z n - 2 May 2003 Biostatistics 6650--L7 9 Interval Estimation: Continuous X Development of 95% CI for .95 Pr( 1.96 1.96) Pr( 1.96 1.96) / Pr( 1.96( / ) 1.96( / )) Pr( 1.96( / ) 1.96( / )) Pr( 1.96( / ) 1.96( / )) Pr( 1.96( / ) Pr( 1....
View Full Document

This note was uploaded on 06/17/2011 for the course BME 6650 taught by Professor Multipleinstructors during the Spring '03 term at Mayo Clinic College of Medicine.

Page1 / 37

Lecture07 - 2 May 2003 Biostatistics 6650--L7 1 Todays...

This preview shows document pages 1 - 10. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online