1 / 9

Chapter 16 – Inference in Practice

Chapter 16 – Inference in Practice. Acceptance Sampling - Accepting or rejecting a sample based on inference ( ex : shipment testing) We still use a Null and Alternate hypothesis: H o : The sample meets standards H a : The sample does not meet standards

kcurrin
Download Presentation

Chapter 16 – Inference in Practice

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. Chapter 16 – Inference in Practice • Acceptance Sampling - Accepting or rejecting a sample based on inference (ex: shipment testing) • We still use a Null and Alternate hypothesis: • Ho: The sample meets standards • Ha: The sample does not meet standards • It is possible that our decision will be wrong: • We may reject a good shipment • We may accept a bad shipment • To distinguish between these two errors, we give them special names: • Type I error: If we reject Ho when in fact Ho is true. • Type II error: If we accept Ho when in fact Ha is true.

  2. @  = 0.05 • We describe the performance of a test by calculating probabilities of Type I and Type II errors… • ex: Shipment of bearings: •  = 2.000 cm •  = 0.010 cm • n = 5 • Ho:  = 2 • Ha:  ≠ 2 • z* = 1.96 • Reject if: • Type I error = rejecting the sample when  really is 2. • Type II error needs a limit - we agree that 2.015 is “bad” - if we were to accept a shipment where  = 2.015, we would be making a Type II error.

  3. The probability of a Type I error =  (0.05 in our example…) • The probability of a Type II error must be calculated… • Step 1: Find ‘acceptance limits’: • Step 2: Calculate the probability of accepting Hoassuming that the alternative is true: • Alt- = 2.015 • Standardize acceptance limits using Alt-:

  4. 2nd VARS -1.39 • So we are looking for the area • between z = -5.32 and -1.39 • So this test: • Will reject 5% of all good samples • Will accept 8.23% of samples bad enough that  = 2.015

  5. Power of a Test • Power - The probability that a fixed level  significance test will reject Ho when alt- is true. (which actually is what should happen…) • aka: “The power of the test”… • Power = p-value of acceptance limit vs. alt-OR • Power = (1 - Type II error) • Essentially says what would happen if we repeated the test many times… • Increasing n ( sample size) will increase the power of the test… • Considerations • Significance testing measures the strength of sample evidence against Ho

  6. Not being able to reject Ho does not mean Ho is true - only that the evidence against it is insufficient … • Decision testing forces a choice between 2 hypotheses - we must say that one or the other is supported by the evidence.

  7. HW 5.4 • Vocab pg 397 • #’s 5.69 - 5.71 all

More Related