stats test #3UPDATED ACTUAL QUESTIONS AND CORRECT Answers

stats test #3UPDATED ACTUAL QUESTIONS AND
CORRECT Answers




uncertainty quantification key takeaways In statistics, you never have to be certain


We don't want to eliminate uncertainty, we want to reduce it when we can and
quantify it always


Where in R can you find uncertainty summary()
under std. error
t value
Pr(>ItI>)

, What does it mean to have heavy and light weight in heavy weight---very likely---small probability---reject
uncertainty terms
little weight--- somewhat likely---reasonable likelihood


steps of hypothesis testing STEP 1:
H0: what we are assuming to be true
HA: What we are trying to prove
Set significance level (𝛼): how much evidence we must have before we reject
H0


STEP #2:
collect sample data (evidence against the assumption)


STEP #3:
summarize the data (weigh the evidence)


STEP #4:
Compare to 𝛼 and make a decision (do we have enough evidence to reject?)


STEP #5:
state conclusion (in terms of HA)


Hypothesis Testing step 1 We deal with H0 and HA, and setting 𝛼


𝛼 is commonly 0.01, 0.05, or 0.1


H0: parameter=value
HA: parameter DOES NOT equal value (x>x>x)


Step 2 of hypothesis testing collect a data sample


evidence against H0


Step 3 of hypothesis testing summarize the data


compute test statistic and p-value


step 4 of hypothesis testing Compare the p-value to 𝛼 and make a decision


p<𝛼 ----- reject hypothesis


Otherwise, don't reject H0


step 5 of hypothesis testing state conclusion for HA


can we conclude that HA is true?


Understanding what a small vs a big p-value means small p-value:
Reject H0 (significant)


Big p-value:
do not reject H0 (not significant)


definition of uncertainty quantification The tool used to quantify the uncertainty in our point estimates goal


(reduce uncertainty and quantify it)