STATISTICAL DATA ANALYSIS IN EXCEL

Microarray Center

STATISTICAL DATA ANALYSIS IN EXCEL

Lecture 6 Some Advanced Topics

Dr. Petr Nazarov

14-01-2013

petr.nazarov@crp-sante.lu

Statistical data analysis in Excel. 6. Some advanced topics

Correction for Multiple Comparisons

Please download the data from edu.data/xls

all_data.xls

Statistical data analysis in Excel. 6. Some advanced topics

2

MULTIPLE EXPERIMENTS

Correct Results and Errors

False Negative, error

False Positive, error

Probability of an error in a multiple test: 1?(0.95)number of comparisons

Statistical data analysis in Excel. 6. Some advanced topics

3

MULTIPLE EXPERIMENTS

False Discovery Rate

False discovery rate (FDR) FDR control is a statistical method used in multiple hypothesis testing to correct for multiple comparisons. In a list of rejected hypotheses, FDR controls the expected proportion of incorrectly rejected null hypotheses (type I errors).

Conclusion

Accept H0 (non-significant) Reject H0 (significant)

Total

Population Condition H0 is TRUE H0 is FALSE

U

T

V

S

m0

m ? m0

Total m ? R

R m

FDR

=

E

V

V +

S

Statistical data analysis in Excel. 6. Some advanced topics

4

MULTIPLE EXPERIMENTS

False Discovery Rate

Assume we need to perform k = 100 comparisons, and select maximum FDR = = 0.05

Statistical data analysis in Excel. 6. Some advanced topics

5

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download