1 / 40

Topic 6 Probability

Topic 6 Probability. Modified from the notes of Professor A. Kuk P&G pp. 125-134. Events: passing an exam getting a disease surviving beyond a certain age treatment effective. An event may occur or may not occur. What is the probability of occurrence of an event?.

Download Presentation

Topic 6 Probability

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Topic 6Probability Modified from the notes of Professor A. Kuk P&G pp. 125-134

  2. Events: • passing an exam • getting a disease • surviving beyond a certain age • treatment effective An event may occuror may not occur. What is the probability of occurrence of an event? Use letters A, B, C, … to denote events

  3. Operations on events 1º Intersection A = “A woman has cervical cancer” B = “Positive Pap smear test” “A woman has cervical cancer and is tested positive”

  4. Venn Diagram S A B

  5. 2° Union • • • • e.g. 6 sided die • • • • • • • • • • • • • • • • • A=“Roll a 3” B=“Roll a 5”

  6. Venn Diagram S A B

  7. 3° Complement “A complement,” denoted by Ac, is the event “not A.” A = “live to be 25” Ac= “do not live to be 25” = “dead by 25”

  8. Venn Diagram S Ac A

  9. Definitions: Null event Cannot happen --- contradiction

  10. Mutually exclusive events: Cannot happen together: A = “live to be 25” B =“die before 10th birthday”

  11. Venn Diagram S B A

  12. Meaning of probability What do we mean when we say P(Head turns up in a coin toss) ? Frequency interpretation of probability Number of tosses 10 100 1000 10000 Proportion of heads .200 .410 .494 .5017

  13. More generally, If an experiment is repeated n times under essentially identical conditions and the event A occurs m times, then as n gets large the ratio approaches the probability of A. as n gets large

  14. For any event A Complement

  15. Venn Diagram Repeat experiment n times Ac=n-m A=m

  16. Mutually exclusive events If A and B are mutually exclusive i.e.cannot occur together

  17. Venn Diagram when A and B are mutually exclusive Conduct experiment n times B=k A=m

  18. Additive Law If the events A, B, C, …. are mutually exclusive – so at most one of them may occur at any one time – then :

  19. In general, B A

  20. Multiplicative rule Note:

  21. Diagnostic tests D = “have disease” Dc =“do not have disease” T+=“positive screening result P(T+|D)=sensitivity P(T-| Dc)=specificity Note: sensitivity & specificity are properties of the test

  22. PRIOR TO TEST P(D)= prevalence AFTER TEST: For someone tested positive, consider P(D|T+)=positive predictive value. For someone tested negative, consider P(Dc |T-)=negative predictive value. Update probability in presence ofadditional information

  23. D T+ Dc

  24. Using multiplicative rule prevalencex sensitivity = prev x sens + (1-prev)x(1-specifity) = positive predictive value = PPV This is called Bayes’ theorem

  25. X-ray Tuberculosis Yes Positive 22 Negative 8 Total 30 Example: X-ray screening for tuberculosis

  26. X-ray Tuberculosis Yes No Positive 22 51 Negative 8 1739 Total 30 1790 Example: X-ray screening for tuberculosis

  27. X-ray Tuberculosis Yes No Positive 22 51 Negative 8 1739 Total 30 1790 Example: X-ray screening for tuberculosis

  28. Screening for TB Population: 1,000,000

  29. Population: 1,000,000 Prevalence = 9.3 per 100,000 No TB: 999,907 TB: 93

  30. Population: 1,000,000 TB: 93No TB: 999,907 = 0.7333 Sensitivity T+ 68 T- 25

  31. Population: 1,000,000 TB: 93No TB: 999,907 Specificity 0.9715 = T+ 68 T- 25 T+ 28,497 T- 971,410

  32. Population: 1,000,000 TB: 93No TB: 999,907 T+ 68 T- 25 T+ 28,497 T- 971,410 T+ 28,565 T- 971,435

  33. Population: 1,000,000 TB: 93No TB: 999,907 T+ 28,497 T+ 68 T+ 28,565 compared with prevalence of 0.00093

  34. Population: 1,000,000 TB: 93No TB: 999,907 T- 971,410 T- 25 T- 971,445

  35. Ingelfinger et.al (1983) Biostatistics in Clinical Medicine

  36. Ingelfinger et.al (1983) Biostatistics in Clinical Medicine

More Related