- CFA Exams
- CFA Level I Exam
- Topic 1. Quantitative Methods
- Learning Module 8. Hypothesis Testing
- Subject 2. Null Hypothesis and Alternative Hypothesis

###
**CFA Practice Question**

An auto manufacturer is conducting a market study for a new model. It is afraid that the new car may cannibalize sales from its existing lineup. The sales director thinks that, based on past experience, as much as 20% or more of existing buyers would move to the new model. The market research department is setting up the null hypothesis to test the effect of the new model on existing model sales. If μ represents the erosion of sales of the current model, which of the following is the most appropriate null hypothesis?

A. μ >= 0%

B. μ = 20%

C. μ >= 20%

**Explanation:**It is already known that the introduction of a new model would cannibalize sales of existing lines, as mentioned by the sales director. He states that the erosion could be 20% or more. Thus, this is the most appropriate null hypothesis. Note that this would make it a one-tailed test.

###
**User Contributed Comments**
6

User |
Comment |
---|---|

mtcfa |
Isn't C the alternative hypothesis, as that is what we are setting out to prove? |

dimanyc |
well, technically, alternative hypothesis should be the "hoped for" answer. so since they don't want any cannibalization, mu<20 is the "hoped for answer". but i think this assumption takes this question 1 extra level though... |

octavianus |
20% or more infers an equality (=, =>, =<) Ho is always stated with the equality, Ha is the opposite of Ho. |

JoeHoong |
I think the answer should be A. The 20% expectation is irrelevant in my opinion. Even a 10% drop would already represent a cannibalization. |

Shaan23 |
I agree with MTCFA. I would think that is the alternate hypothesis as well. That is what we are trying to prove. That option didnt exist though but if it did I would choose that. |

edrei7 |
C is correct. The null hypothesis is the "status quo". We are trying to find out from the data gathering if the results are significant enough that less than 20% of existing buyers will move to the new model. That is what we are trying to prove. Because from past experience, we already expect that 20% or more of existing buyers will move to the new model. |