1 / 25

Errors in Hypothesis Testing

Errors in Hypothesis Testing. 2 TYPES OF ERRORS. TRUE CASE H A is true H A is false WE Accept H A SAY Do not Accept H A. TYPE I ERROR. CORRECT. PROB = α. TYPE II ERROR. CORRECT. PROB = β. α is set by the decision maker. β varies and depends on:

bryant
Download Presentation

Errors in Hypothesis Testing

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. Errors in Hypothesis Testing

  2. 2 TYPES OF ERRORS TRUE CASE HAis true HAis false WE Accept HA SAY Do not Accept HA TYPE I ERROR CORRECT PROB = α TYPE II ERROR CORRECT PROB = β α is set by the decision maker β varies and depends on: (1) α; (2) n; (3) the true value of 

  3. Relationship Between  and  •  is the Probability of making a Type II error • i.e. the probability of not concluding HA is true when it is •  depends on the true value of  • The closer the true value of  is to its hypothesized value, the more likely we are of not concluding that HA is true -- i.e.  is large (closer to 1) •  is calculated BEFOREa sample is taken • We do not use the results of a sample to calculate 

  4. The Hypothesis Test CALCULATING  • Example: If we take a sample of n = 49, with  = 4.2, “What is the probability we will get a sample from which we would not conclude  > 25 when  really = 25.5?” (Use  = .05) REWRITE REJECTION REGION IN TERMS OF

  5. That’s a TYPE II ERROR!! P(Making this error) =  CALCULATING  (cont’d) • So when  = 25.5, • If we get an > 25.987, we will correctly conclude that  > 25 • If we get an < 25.987 we will not conclude that  > 25 even though  really = 25.5

  6. β CALCULATING  (cont’d) • So what is P(not getting an > 25.987 when  really = 25.5? That is P(getting an < 25.987)? Calculate z = (25.987 - 25.5)/(4.2/ )  .81 •  is the area to theleftof .81 for a “>” test • P(Z < .81) = .7910

  7. DO NOT ACCEPT HA WRONG Prob = =.7910 ACCEPT HA RIGHT! “>” TestDetermining  When  = 25.5 .7910 25.5 25.987 0 .81 Z

  8. β What is  When  = 27? • So what is P(not getting an > 25.987 when  really = 27? That is P(getting an < 25.987)? Calculate z = (25.987 - 27)/(4.2/ )  -1.69 •  is the area to theleftof -1.69 for a “>” test • P(Z < -1.69) = .0455 This shows that the further the true value of  is from the hypothesized value of , the smallerthe value of β; that is we are less likely to NOT conclude that HA is true (and it is!)

  9. DO NOT ACCEPT HA WRONG Prob = =.0455 ACCEPT HA RIGHT! “>” TestDetermining  When  = 27 .0455 25.987 27 -1.69 0 Z

  10. The Hypothesis Test  for “<” Tests • For n = 49,  = 4.2, “What is the probability of not concluding that  < 27, when  really is 25.5? (With  = .05) • This time  is the area to the right of

  11. β What is  When  = 25.5? • So what is P(not getting an < 26.013 when  really = 25.5? That is P(getting an > 26.013)? Calculate z = (26.013 – 25.5)/(4.2/ )  .86 •  is the area to therightof .86 for a “<” test • P(Z > .86) = 1 - .8051 = .1949

  12. DO NOT ACCEPT HA WRONG Prob = =.1949 ACCEPT HA RIGHT! .1949 “<” TESTDetermining  When  = 25.5 .8051 25.5 26.013 0 .86 Z

  13. The Hypothesis Test  for “” Tests • For n = 49,  = 4.2, “What is the probability of not concluding that   26, when  really is 25.5? (With  = .05) • This time  is the area in the middle between the two critical values of

  14. β What is  When  = 25.5? • So what is P(not getting an < 24.824 or > 27.176 when  really = 25.5? That is P(24.824 < < 27.176)? Calculate z’s = (24.824 – 25.5)/(4.2/ )  -1.13 and = (27.176 – 25.5)/(4.2/ )  2.79 •  is the areain between -1.13 and 2.79 for a “” test • P(Z < 2.79) = .9974 • P(Z < -1.13) = .1292 P(-1.13 < Z < 2.79 = .9974 - .1292 = .8682

  15. DO NOT ACCEPT HA WRONG Prob = =.9974 – .1292 =.8682 ACCEPT HA RIGHT! .8682 .9974 .1292 “” TESTDetermining  When  = 25.5 24.824 25.5 27.176 -1.13 0 2.79 Z

  16. The Power of a Test = 1 -  •  is the Probability of making a Type II error • i.e. the probability of not concluding HA is true when it is •  depends on the true value of  and sample size, n • The Power of the test for a particular value of  is defined to be the probability of concluding HA is true when it is -- i.e. 1 - 

  17. Power Curve Characteristics • The power increases with: • Sample Size, n • The distance the true value of μ is from the hypothesized value of μ

  18. n = 49 n = 25 α = .05 Power Curves For HA: μ 26With n = 25 and n = 49

  19. Calculating  Using Excel“> Tests” Suppose H0 is  = 25;  = 4.2, n = 49,  = .05 “>” TESTS: HA:  > 25 and we want  when the true value of  = 25.5 1) Calculate the criticalx-bar value = 25 + NORMSINV(.95)*(4.2/SQRT(49)) 2) Calculate z =(criticalx-bar -25.5)/ (4.2/SQRT(49)) 3) Calculate the the probability of getting a z- value < than this critical z value: -- this is =NORMSDIST(z)

  20. Calculating  Using Excel“< Tests” Suppose H0 is  = 27;  = 4.2, n = 49,  = .05 “< TESTS”: HA:  < 27 and we want  when the true value of  = 25.5 1) Calculate the critical x-bar value = 27 - NORMSINV(.95)*(4.2/SQRT(49)) 2) Calculate z =(criticalx-bar -25.5)/ (4.2/SQRT(49)) 3) Calculate the the probability of getting a z- value > than the critical value: -- this is  =1-NORMSDIST(z)

  21. Calculating  Using Excel“ Tests” Suppose H0 is  = 26;  = 4.2, n = 49,  = .05  TESTS: HA:   26 and we want  when the true value of  = 25.5 1) Calculate the critical upperx-barU value and the lower criticalx-barL value = 26 - NORMSINV(.975)*(4.2/SQRT(49)) (x-barL) = 26 + NORMSINV(.975)*(4.2/SQRT(49)) (x-barU) 2) Calculate zU=(x-barU-25.5)/ (4.2/SQRT(49)) and zL=(x-barL-25.5)/ (4.2/SQRT(49)) 3) Calculate the the probability of getting an z- value in between zL and zU - this is =NORMSDIST(zU) - NORMSDIST(zL)

  22. =B3+NORMSINV(1-B2)*(B5/SQRT(B6)) =(B8-B7)/(B5/SQRT(B6)) =NORMSDIST(B9) =1-B10 β for “>” Tests

  23. =B3-NORMSINV(1-B2)*(B5/SQRT(B6)) =(B8-B7)/(B5/SQRT(B6)) =1-NORMSDIST(B9) =1-B10 β for “<” Tests

  24. =B3-NORMSINV(1-B2/2)*(B5/SQRT(B6)) =B3+NORMSINV(1-B2/2)*(B5/SQRT(B6)) =(B8-B7)/(B5/SQRT(B6)) =(B9-B7)/(B5/SQRT(B6)) =NORMSDIST(B11)-NORMSDIST(B10) =1-B12 β for “” Tests

  25. REVIEW • Type I and Type II Errors •  = Prob (Type I error) •  = Prob (Type II error) -- depends on , n and α • How to calculate  for: • “>” Tests • “<” Tests • “” Tests • Power of a Test at  = 1-  • How to calculate  using EXCEL

More Related