Chapter 4b. Basic statistical tests for inference

Chapter 4b. Basic statistical tests for inference - Basic...

Info iconThis preview shows pages 1–7. Sign up to view the full content.

View Full Document Right Arrow Icon
1 Basic statistical tests for inference
Background image of page 1

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

View Full DocumentRight Arrow Icon
2 ± Confidence interval ± Comparing means ± Assessing normality ± Relationship between 2 continuous variables ± Relationship between 2 categorical variables
Background image of page 2
3 Statistical inference ± In addition to descriptive statistics, PROC UNIVARIATE generates output for inference such as: ² One sample t-test, sign-test and signed rank test; (the latter 2 tests can be regarded as non-parametric form of the one sample t-test) ² Confidence intervals for the mean and variance (assuming the underlying population is normally distributed) ² Formal tests for assessing whether the distribution is a normal distribution or not ± These are requested through the PROC UNIVARIATE statements options ² MU0, CIBASIC, ALPHA, NORMAL
Background image of page 3

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

View Full DocumentRight Arrow Icon
4 Confidence interval ± With options CIBASIC and ALPHA, 99% confidence intervals for the mean, standard deviation and variance for the variable in additional to the output * example 4b.1 proc univariate example; * generating a 99% confidence interval; proc univariate data=eg4b.prog cibasic alpha=0.01; var progdays; where (year=1); run; Basic Confidence Limits Assuming Normality Parameter Estimate 99% Confidence Limits Mean 142.12500 35.61342 248.63658 Std Deviation 86.08706 50.57980 228.99849 Variance 7411 2558 52440
Background image of page 4
5 ± Confidence intervals for the mean can be obtained from PROC MEANS through option CLM and ALPHA * example 4b.2 proc means example; * generating a 99% confidence interval; proc means data=eg4b.prog clm alpha=0.01; var progdays; where (year=1); run; Analysis Variable : progdays Lower 99% Upper 99% CL for Mean CL for Mean ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ 35.6134230 248.6365770 ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ
Background image of page 5

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

View Full DocumentRight Arrow Icon
6 Comparing means ± With option MU0, a test is performed on the mean using a one-sample t-test ² H 0 : μ = 0 vs H 1 : 0 * example 4b.3 proc univariate example; * perform a 1-sample t-test; proc univariate data=eg4b.prog mu0=140; var progdays; where (year=1); run;
Background image of page 6
Image of page 7
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 02/09/2012 for the course STAT 1301 taught by Professor Smslee during the Spring '08 term at HKU.

Page1 / 22

Chapter 4b. Basic statistical tests for inference - Basic...

This preview shows document pages 1 - 7. Sign up to view the full document.

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