Lecture15 - 26 June 2003 Biostatistics 6650-L15 1 Todays...

Info iconThis preview shows pages 1–8. Sign up to view the full content.

View Full Document Right Arrow Icon
26 June 2003 Biostatistics 6650--L15 1
Background image of page 1

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

View Full DocumentRight Arrow Icon
26 June 2003 Biostatistics 6650--L15 2 Today’s Schedule Association of continuous X with continuous Y brief introduction Scatter plots Correlation Pearson/Spearman Regression simple linear regression least squares Wrap-up
Background image of page 2
26 June 2003 Biostatistics 6650--L15 3 Scatter Plots Before doing any calculations---PLOT the DATA Assume Y and X are two continuous variables of interest both are measured on each subject one is not calculated from the other data arrangement N subjects 2 columns Y and X JMP: analyze Fit Y*X or analyze multivariate specify two continuous variables default output is a scatter plot
Background image of page 3

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

View Full DocumentRight Arrow Icon
26 June 2003 Biostatistics 6650--L15 4 Scatter Plots Lab 3 IgA Nephropathy baseline data, N=106 X=age, Y=systolic blood pressure 80 90 100 110 120 130 140 150 160 170 180 SBP_0 10 20 30 40 50 60 70 80 AGE_RAN Bivariate Fit of SBP_0 By AGE_RAN
Background image of page 4
26 June 2003 Biostatistics 6650--L15 5 Correlation • Covariance= E{(X-6 x )(Y-6 y )} measures how X and Y vary together or co-vary if X and Y are independent, – then Covariance= E(X-6 x )*E(Y-6 y )=0*0=0 if X and Y are positively related, then when X is high relative to its mean, Y will also be high. This leads to large positive covariance. Population Correlation coefficient(6 6 = standardized covariance = E{(X-6 x )(Y-6 y )} / (6 x 6 y ), measures strength of linear association between Y and X ranges from -1 (perfect negative association) to 0 (no association) to 1 (perfect positive association) Caution: combining groups with different 6’s may produce spurious results
Background image of page 5

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

View Full DocumentRight Arrow Icon
26 June 2003 Biostatistics 6650--L15 6 Correlations Fig 5.16 Rosner -2.5 0 2.5 5 7.5 10 y 0 2 4 6 8 x Bivariate Fit of y By x i =0, but perfect non-linear relationship between Y and X---plot the data!
Background image of page 6
26 June 2003 Biostatistics 6650--L15 7 Correlation Sample estimate of 6, is r(Pearson’s r) defined as: • Spearman’s rank correlation(r s ) Calculate Pearson’s r using the ranks of X and Y. Less affected by outliers
Background image of page 7

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

View Full DocumentRight Arrow Icon
Image of page 8
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 19

Lecture15 - 26 June 2003 Biostatistics 6650-L15 1 Todays...

This preview shows document pages 1 - 8. Sign up to view the full document.

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