In: Statistics and Probability
Answer TRUE or FALSE for each of these
statements.
Multiple comparison testing using the Bonferroni correction is only
applied if we reject the null hypothesis in a one-way ANOVA.
TRUEFALSE [1 pt(s)] Tries 0/1In the rank-sum test, when the
magnitude of the test-statistic gets larger, the p-value gets
smaller. TRUEFALSE [1 pt(s)] Tries 0/1A p-value of 0.001 indicates
we should accept the alternative hypthesis. TRUEFALSE [1 pt(s)]
Tries 0/1A regression model that is statistically significant
according to an F-test establishes a strong case for a cause and
effect relationship. TRUEFALSE [1 pt(s)] Tries 0/1SSR (Sum of
Squares Regression) is based on the squared discrepancies between
the observations and the predicted responses. TRUEFALSE [1 pt(s)]
Tries 0/1If height is measured in metres, the units for the
variance of height is also measured in metres. TRUEFALSE [1 pt(s)]
Tries 0/1In a pooled two-sample t-test, the assumption of equal
variance is violated if the sample standard deviation for the first
group is different than the sample standard deviation for the
second group. TRUEFALSE [1 pt(s)] Tries 0/1A 95% confidence
interval is the same as a hypothesis test with a two-sided
alternative at significance level α = 0.05. TRUEFALSE [1 pt(s)]
Tries 0/1Non-parametric statistical tests are used when the
independence assumption is violated. TRUEFALSE [1 pt(s)] Tries 0/1A
residual plot allows us to check all the assumptions in a
regression model. TRUEFALSE [1 pt(s)] Tries 0/1In simple linear
regression, the absolute value of the correlation coefficient is
always as large or larger than R2. TRUEFALSE [1
pt(s)] Tries 0/1If R2=1 in a regression, then SSE
is also 1. TRUEFALSE [1 pt(s)] Tries 0/1
Answer TRUE or FALSE for each of these statements.
## Multiple comparison testing using the Bonferroni correction is only applied if we reject the null hypothesis in a one-way ANOVA.
Answer : True ( comparison test is valid only if we reject Ho , from test we can find out which mean or pair is differ)
## In the rank-sum test, when the magnitude of the test-statistic gets larger, the p-value gets smaller.
Answer : True
## A p-value of 0.001 indicates we should accept the alternative hypthesis.
Answer : True : ( we reject Ho if p value is less than alpha value here p value is 0.001 which is less than for 5% , 1%, and 10 % level of significance . and its commonly used hence for it if p value is less than alpha we reject Ho and Accept H1 ( alternative )
## A regression model that is statistically significant according to an F-test establishes a strong case for a cause and effect relationship.
Answer : True
## SSR (Sum of Squares Regression) is based on the squared discrepancies between the observations and the predicted responses.
Answer : False ( it is SSE =sum of squre of error )
## if height is measured in metres, the units for the variance of height is also measured in metres.
Answer : False ( variance should meausred in meters^2 )
## In a pooled two-sample t-test, the assumption of equal variance is violated if the sample standard deviation for the first group is different than the sample standard deviation for the second group.
Answer : False ( it is violated if population standard deviation for the first group is different than the population standard deviation for the second group )
## A 95% confidence interval is the same as a hypothesis test with a two-sided alternative at significance level α = 0.05.
Answer : True
## Non-parametric statistical tests are used when the independence assumption is violated.
Answer : True
## A residual plot allows us to check all the assumptions in a regression model.
Answer : True ( check constant variance or not )
## In simple linear regression, the absolute value of the correlation coefficient is always as large or larger than R2.
Answer : True
## If R2=1 in a regression, then SSE is also 1.
Answer = false ( SST = SSE must not only SSE 1 )