Statistics for Decision- Making in Business - Maricopa Community ...
Statistics for Decision- Making in Business - Maricopa Community ...
Statistics for Decision- Making in Business - Maricopa Community ...
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
2) <strong>Decision</strong> Rule: We will reject the null hypothesis when the likelihood of<br />
observ<strong>in</strong>g someth<strong>in</strong>g as small/large or smaller/larger than 16 out of 1000<br />
bushels is no larger than a 1% probability, under the assumption of the null<br />
hypothesis. That is,<br />
( )<br />
3) We will reject if the observed value of is smaller or larger than some<br />
cutoff values of . That is, if it is smaller than some value, say , or larger than<br />
some value, say , then we will reject . Remember, we set-up a hypothesis<br />
first, then do the test. Even though 16 is larger than 15 out of 1000, we did not<br />
know this to beg<strong>in</strong> with. We are still test<strong>in</strong>g whether or not this value is<br />
significantly different and do not care about the direction of the difference.<br />
4) Based on the sample evidence, we will either:<br />
a) Reject <strong>in</strong> favor of of mach<strong>in</strong>es fail. That is, either a<br />
significantly fewer number of them fail, or a significantly greater number<br />
of them fail.<br />
b) Fail to reject . We do not have sufficient evidence to conclude that new<br />
mach<strong>in</strong>es fail more or less when compared to the old mach<strong>in</strong>e.<br />
10.<br />
1) Type I: We conclude the farmer‟s method reduces crop destruction, when there is<br />
no difference; Type II: We conclude the farmer‟s method is no different than the<br />
old method, when <strong>in</strong> fact there is less than 7% crop destruction with his new<br />
method.<br />
2) Type I: We conclude the <strong>in</strong>structors students per<strong>for</strong>m better than his <strong>for</strong>mer<br />
students, when <strong>in</strong> fact there is no difference; Type II: We conclude that his new<br />
students per<strong>for</strong>m just as well as his <strong>for</strong>mer students, when <strong>in</strong> fact they do better.<br />
3) Type I: We conclude that the new mach<strong>in</strong>es fail more or less than the <strong>for</strong>mer<br />
mach<strong>in</strong>es, when <strong>in</strong> fact there is no difference; Type II: We conclude that there is<br />
no difference between the failure rates of the new and old mach<strong>in</strong>es, when <strong>in</strong> fact<br />
there is a significant difference.<br />
11. Increas<strong>in</strong>g means we will reject less often, as we set more str<strong>in</strong>gent conditions upon<br />
the rejection process. If we reject less often, then there is an elevated likelihood that we<br />
may fail to reject, when <strong>in</strong> fact we should. This is precisely what a Type II error is.<br />
<strong>Statistics</strong> <strong>for</strong> <strong>Decision</strong>-<strong>Mak<strong>in</strong>g</strong> <strong>in</strong> Bus<strong>in</strong>ess © Milos Podmanik Page 246