MGEB12H3 Chapter Notes - Chapter 9: Null Hypothesis, Alternative Hypothesis, Statistical Parameter

Economics for Management Studies
Course Code
Ataollah Mazaheri

This preview shows half of the first page. to view the full 2 pages of the document.
Hypothesis Testing: can be used to determine whether a statement about the value of a population
parameter should or should not be rejected.
- Null hypothesis
- Alternative hypothesis
Usually set up to see if enough evidence can reject the null while to accept the alternative.
Answering in statistics world:
- NEVER actually prove the null or alternative based on statistics, only have enough evidence to
support the null/alternative.
o HOWEVER, there is always a chance of error, even with enough evidence.
- ONLY conclude or object alternative
o ONLY If evidence from sample is strong enough to reject the null, you accept the
o Other-wise, you do not accept alternative, but do not prove null, only strong evidence to
support null.
Null hypothesis: H0, is a tentative assumption about a population parameter.
- Easiest way to identify a null: = sign ie. More or equal, or, less or equal
Alternative hypothesis: Ha, is the opposite of what is stated in the null hypothesis.
Usually the research hypothesis should be expressed as the alternative hypothesis.
- Thus, if evidence is high to support the research hypothesis: null is rejected while alternative is
i.e. testing the validity of a claim
- claim: null
- usually test for the alternative, so if high enough we can accept alternative and reject the
null or conclude with unknown/inconclusive
Null and Alternative Hypothesis about a Population Mean, µ
- µ0 is the hypothesized value of the population mean
*just write = for H0 redundant if include less or more than
*assume the null is accepted at =
- Attempt to reject null when assuming it is correct
o Cannot do a formal test if null is not assumed correct
find more resources at
find more resources at
You're Reading a Preview

Unlock to view full version