Confidence interval 1 sample for mean(σ is unknown)
s s
Two side: x−t α / 2 ,n−1 ≤ μ ≤ x +t α /2 , n−1
√n √n
The distribution of Revenue of an ice cream business has a normal distribution. A 50 random sample of
revenue has the following amount(in dollars)
Construct a 95% confidence interval for the true average Revenue of that ice cream business.
Confidence interval 1 sample for proportion
Two side: ^p−z ∝/ 2
√ ^p (1− ^p )
n √
≤ p ≤ ^p + z ∝/ 2
^p (1− ^p ) ^ x
n
( P= ¿
n
Of 50 randomly selected Revenue of a ice cream business, 16 resulted in Revenue greater than 600$.
Calculate a 95% two-sided confidence interval of the Revenue rate that greater than 600 dollars.
Test of hypothesis 1 sample for mean(σ is unknown)
Tradition method:
Step 1: Form the two hypotheses H0: 𝜇 = 𝜇0 and H1: 𝜇 ≠ 𝜇0
x−μ0
Step 2: Compute the test statistic: t 0=
s/√n
Step 3: Identify acceptance region (−t a ,n−1 ; t a , n−1) , use t-distribution with df = n-1
2 2
Step 4: Make a decision:
If the test statistic t0 is in critical region, then reject H0
If the test statistic t0 is in acceptance region, then fail to reject H0
The Revenue of an ice cream business has a normal distribution. A 50 random revenue sample are
collected and has the following price(in dollars):
Test the hypothesis that the average The Revenue of a ice cream business is 600$, alpha = 0.05.
Test of hypothesis 1 sample for proportion
Tradition method:
• Step 1: Construct the two hypotheses H0: p = p0 and H1 : p ≠ p0
• Step 2: Find the test statistic:
• Step 3: Identify acceptance region, use Z = N(0,1).
• Step 4: Make a decision:
If the test statistic is in critical region, then reject H0
If the test statistic is in acceptance region, then fail to reject H0
A ice cream business claims that percentage of the revenue that is greater than 600$ is 50% .In a random
sample of 50 revenue it is found out that 16 of them are greater than 600$. Test the claim of a ice cream
business at the significance level of 5%.
Confidence interval on the difference of 2 means
Two-side:
+)Assume equal variances (x ¿ ¿ 1−x 2)−t α/2 , df
√ s 2p s2p
√
+ ≤ μ −μ ≤( x ¿ ¿ 1−x 2 )+ t α /2 , df
n1 n 2 1 2
s 2p s 2p
+ ¿¿
n1 n2
+)Not assume equal variances(x ¿ ¿ 1−x 2)−t α/2 , df
√ s 21 s 22
√
+ ≤ μ −μ ≤( x ¿ ¿ 1−x 2 )+ t α /2 , df
n1 n 2 1 2
s21 s 22
+ ¿¿
n1 n2
Given dataset below:
Construct 95% confidence interval for difference in means in two case :
+)variances unknown (assume equal variances)
+)variances unknown (not assume equal variances)
Test of hypothesis 2 sample for mean
+)Assume equal variances:
Step 1: Construct the two hypotheses
H0: μ1−μ2 =∆0
H1 : μ1−μ2 ≠ ∆0
Step 2: Find the test statistic:
( x 1−x 2) −∆ 0
t 0=
√
2 2
sp s p
+
n 1 n2
Step 3: Identify acceptance region, use t-distribution with df = n1+n2-2.
Step 4: Make a decision:
If the test statistic is in critical region, then reject H0
If the test statistic is in acceptance region, then fail to reject H0
+)Not assume equal variances:
Step 1: Construct the two hypotheses H0: μ1−μ2 =∆0 , H1 : μ1−μ2 ≠ ∆0
Step 2: Find the test statistic:
( x 1−x 2) −∆ 0
t 0=
√
2 2
s 1 s2
+
n 1 n2
Step 3: Identify acceptance region, use t-distribution with degree of freedom df .
Step 4: Make a decision:
If the test statistic is in critical region, then reject H0
If the test statistic is in acceptance region, then fail to reject H0
Continue example above:
+)Use significant level 0.05 and assume equal variances, is there any difference in the mean yields?.
+)Use significant level 0.05 and assume equal variances, is there any difference in the mean yields?.
Confidence interval on the difference of 2 proportions:
confidence interval for ( p1− p2 ¿ is:
( ^p1− ^p 2 ¿−z α / 2
√ ^p1 (1−^p 1) ^p2 (1−^p 2)
n1
+
n2 √
≤ p1− p2 ≤ ( ^p1− ^p2 ¿+ z α / 2
^p 1(1−^p1) ^p 2(1−^p2)
n1
+
n2
50 Revenue of an ice cream business are taken to analyzed as sample 1. 70 Revenue of an ice cream
business are taken to analyzed as sample 2. The number of Revenue in sample 1 that greater than 600$
is 16, and sample 2 is 22.
Construct 95% confidence interval for difference of two these proportions.
Test of hypotheses on the difference of 2 proportions:
Step 1: Construct the two hypotheses H0: p1− p2=0 , H1: p1− p2 ≠ 0
Step 2: Find the test statistic:
z 0=( ^p 1−^p 2 ¿ ¿
√ p(1−p) p (1− p)
n1
+
n2
Step 3: Identify acceptance region, use Z = N(0,1).
Step 4: Make a decision:
If the test statistic is in critical region, then reject H0
If the test statistic is in acceptance region, then fail to reject H0
Regression Prediction
^y = β^ 0 + ^β 1x
The point estimates of β 0, β 1 , say ^β 0, ^β 1 are:
S xy ^
^β = ; β = y − ^β1 x ,
1
S xx 0
(∑ x i )
2
S xx =∑ ( x i−x ) =∑ xi −
2 2
n
( ∑ x i)( ∑ y i )
S xy =∑ ( x i−x ) ( y i− y )=∑ x i y i−
n
Regression Test hypothesis about the slope and intercept
• Estimated of regression slope β 1 is ^β 1
• Estimated of regression intercept β 0 is ^β 0
• Estimated standard error of the slope is se ( β^ 1 ) =
√ σ^ 2
S xx
• Estimated standard error of the intercept is
√ 1 x2
se ( β^ 0 ) = σ^ 2 ( + )
n S xx
Total sum of squares:
(∑ y i )
2
SST =∑ ( y i− y ) =∑ y i −
2 2
n
Regression sum of squares:
SS R=∑ ( ^y i− y ) = β^ 1 S xy
2
Error sum of squares:
SS E=∑ ( y i−^y i ) =SST −SS R
2
An unbiased estimator of σ 2 is
2 SS E
σ^ =
n−2
