session 4: samples and sampling distributionspopulation sample1 sample2 sample3 mean 0.500 0.504...
TRANSCRIPT
![Page 1: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/1.jpg)
Session 4: Samples and samplingdistributions
Susan Thomas
http://www.igidr.ac.in/˜susant
IGIDR
Bombay
Session 4: Samples and sampling distributions – p. 1
![Page 2: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/2.jpg)
Recap
Probability and principles
Random variables
Probability distributions and densities
Parameters of probability distributions and densities
Data descriptors
Session 4: Samples and sampling distributions – p. 2
![Page 3: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/3.jpg)
The aim of this session
1. Samples and populations
2. Parameters vs. statistics
3. Sampling distribution of statistics
4. Central Limit Theorem
5. Estimating the population mean
Session 4: Samples and sampling distributions – p. 3
![Page 4: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/4.jpg)
Samples and populations
Session 4: Samples and sampling distributions – p. 4
![Page 5: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/5.jpg)
Samples
A sample is a subset of the possible values for a RV.
All the possible values is called the population.
Samples are available; populations are not.
Samples are used to infer characteristics of thepopulation.
A good sample is representative of the population,in terms of1. The values2. The frequency of the values
Session 4: Samples and sampling distributions – p. 5
![Page 6: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/6.jpg)
Example from a uniform distribu-tion, u
The population is U(0,1).
u = Sample of size 25
In R, the commandu = runif(25)generates a sample of 25 draws from U(0,1)
Session 4: Samples and sampling distributions – p. 6
![Page 7: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/7.jpg)
Histogram of u
Histogram of x
x
Den
sity
0.0 0.2 0.4 0.6 0.8 1.0
0.0
0.5
1.0
1.5
2.0
Session 4: Samples and sampling distributions – p. 7
![Page 8: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/8.jpg)
Histogram of u
Histogram of x$V1
x$V1
Den
sity
0.0 0.2 0.4 0.6 0.8 1.0
0.0
0.5
1.0
1.5
2.0
Session 4: Samples and sampling distributions – p. 8
![Page 9: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/9.jpg)
Summary statistics
Population Sample
Mean 1.000 0.504Sigma 0.289 0.2941st Quartile 0.250 0.2753rd Quartile 0.750 0.792
Session 4: Samples and sampling distributions – p. 9
![Page 10: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/10.jpg)
Second sample from U(0,1) u2
The population is U(0,1).
u2 = Sample of size 25
In R, the commandu2 = runif(25)generates a second sample of 25 draws from U(0,1)
Session 4: Samples and sampling distributions – p. 10
![Page 11: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/11.jpg)
Summary statistics
Population Sample1 Sample2
Mean 1.000 0.504 0.422Sigma 0.289 0.294 0.2031st Quartile 0.250 0.275 0.3113rd Quartile 0.750 0.792 0.534
Session 4: Samples and sampling distributions – p. 11
![Page 12: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/12.jpg)
Third sample from U(0,1) u3
The population is U(0,1).
u2 = Sample of size 2500
In R, the commandu2 = runif(2500)generates a second sample of 2500 draws fromU(0,1)
Session 4: Samples and sampling distributions – p. 12
![Page 13: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/13.jpg)
Summary statistics
Population Sample1 Sample2 Sample3
Mean 0.500 0.504 0.422 0.494Sigma 0.289 0.294 0.203 0.2891st Quartile 0.250 0.275 0.311 0.2373rd Quartile 0.750 0.792 0.534 0.741
Session 4: Samples and sampling distributions – p. 13
![Page 14: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/14.jpg)
Statistics vs. Parameters
A numerical descriptive measure of a population iscalled a parameter.For example, the bernoulli distribution has one parameter p,and the normal distribution has two parameters µ, σ.
A quantity calculated from a sample set ofobservations of the RV is called a statistic.
Parameters of a given distribution are constant.Statistics calculated for different samples from thesame distribution are different – they are randomvariables!For example, the mean of a sample is a RV is a randomvariable.
Session 4: Samples and sampling distributions – p. 14
![Page 15: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/15.jpg)
Sampling distributions
Session 4: Samples and sampling distributions – p. 15
![Page 16: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/16.jpg)
Sampling distribution of samplestatistics
Like all RVs, sample statistics have probabilitydistributions – these are called samplingdistributions.
The sampling distribution is the relative frequencydistribution, theoretically generated by1. Repeatedly taking random samples (of size n) of the RV,
2. Calculating the statistic for each sample
Each sample used in generating the samplingdistribution has to have the same number ofobservations.
Therefore, the sampling distribution is generated for(a) a given statistic and (b) for a given sample size.
Session 4: Samples and sampling distributions – p. 16
![Page 17: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/17.jpg)
Examples of the samplingdistribution of sample means, x̄
Session 4: Samples and sampling distributions – p. 17
![Page 18: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/18.jpg)
Example1: n = 25, U(0,1)
We will create 100 samples to generate the samplingdistribution of the mean from U(0,1), n = 251. Sample 1, S1 = runif(25)2. Mean 1, m[1] = mean(S1)
Run 1. and 2. in a loop to get 100 values for themean stored in m.
Session 4: Samples and sampling distributions – p. 18
![Page 19: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/19.jpg)
Analysing the generated data
Visually:1. To get the histogram of the means:
hist(m,breaks=20,freq=FALSE,col=5)2. To superimpose the kernel density plot over the
histogram:lines(density(m),col=2,lwd=4)
Numerically:
1. To get the mean, 1st and 3rd quartile:summary(m)
Session 4: Samples and sampling distributions – p. 19
![Page 20: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/20.jpg)
Example1: histogram of x̄ for U(0,1),n = 25
Histogram of mus
mus
Den
sity
0.35 0.40 0.45 0.50 0.55 0.60 0.65
02
46
8
Session 4: Samples and sampling distributions – p. 20
![Page 21: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/21.jpg)
Example1: Summary statistics of x̄
x̄
Mean 0.494Standard deviation 0.0611st quartile 0.4533rd quartile 0.533
Session 4: Samples and sampling distributions – p. 21
![Page 22: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/22.jpg)
Example2: n = 25, N(0,1)
We will create 100 samples to generate the samplingdistribution of the mean from N(0,1), n = 251. Sample 1, S1 = rnorm(25)2. Mean 1, m[1] = mean(S1)
Run 1. and 2. in a loop to get 100 values for themean stored in m.
Session 4: Samples and sampling distributions – p. 22
![Page 23: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/23.jpg)
Example2: histogram of x̄ of N(0,1),n = 25
Histogram of mus
mus
Den
sity
−0.5 0.0 0.5
0.0
0.5
1.0
1.5
2.0
Session 4: Samples and sampling distributions – p. 23
![Page 24: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/24.jpg)
Example2: Summary statistics of x̄
x̄
Mean 0.003Standard deviation 0.1991st quartile -0.1313rd quartile 0.136
Session 4: Samples and sampling distributions – p. 24
![Page 25: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/25.jpg)
Central Limit Theorem
Session 4: Samples and sampling distributions – p. 25
![Page 26: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/26.jpg)
Stating the CLT
If the sample size is sufficiently large,
the sample mean x̄ has a sampling distribution thatis approximately normal
This is irrespective of the distribution of theunderlying RV.
Session 4: Samples and sampling distributions – p. 26
![Page 27: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/27.jpg)
Properties of the sampling distribu-tion of x̄
If x̄ is the mean of a sample of RVs from a populationwith mean parameter µ and standard deviation parameterσ, then:
1. The mean of the sampling distribution of x̄ is calledµx̄ and is equal to population µ.
µx̄ = µ
2. The standard deviation of the sampling distributionof x̄ is called σx̄ and is a fraction of the populationσ, as:
σx̄ = σ/√
n
Where n is the size of the sample. Session 4: Samples and sampling distributions – p. 27
![Page 28: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/28.jpg)
Recap
1. PopulationParameters: µ, σ
This distribution generates the underlying RV.
2. Sample of size n:Statistics: x̄, σ̄
This is a feature of one sample.
3. Distribution of sample statistics from samples ofsize n
Sampling distribution parameters: µx̄, σx̄
This is a frequency distribution from a set of samples all
the same size.Session 4: Samples and sampling distributions – p. 28
![Page 29: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/29.jpg)
Testing concepts
A small-town newspaper reported that for families in their
circulation area, the distribution of weekly expenses for food
consumed away from home has an average of Rs.237.60 and a
standard deviation of Rs.50.40. An economist randomly sampled
100 families for their outside-home food expenses for a week.
1. What is the distribution of the mean weekly outside-home food
expenses for the 100 families?
It should be approximately
normal distributed, with µx̄ = Rs.237.60, and
σ = 50.40/√
100 = 5.50
2. What is the probability that the sample mean weekly expenses
will be at least Rs.252?
We convert N(0,1) distribution as
z = 252 − 237.60/5.5 = 2.62. Then Pr(average expenses >
252) = Pr(z > 2.62) = 0.5 − 0.4956 = 0.0044%
Session 4: Samples and sampling distributions – p. 29
![Page 30: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/30.jpg)
Testing concepts
A small-town newspaper reported that for families in their
circulation area, the distribution of weekly expenses for food
consumed away from home has an average of Rs.237.60 and a
standard deviation of Rs.50.40. An economist randomly sampled
100 families for their outside-home food expenses for a week.
1. What is the distribution of the mean weekly outside-home food
expenses for the 100 families? It should be approximately
normal distributed, with µx̄ = Rs.237.60, and
σ = 50.40/√
100 = 5.50
2. What is the probability that the sample mean weekly expenses
will be at least Rs.252? We convert N(0,1) distribution as
z = 252 − 237.60/5.5 = 2.62. Then Pr(average expenses >
252) = Pr(z > 2.62) = 0.5 − 0.4956 = 0.0044%Session 4: Samples and sampling distributions – p. 29
![Page 31: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/31.jpg)
Testing concepts: Comparing sam-pling distributions
Say x̄25 is the mean of a random sample of size 25 from a population
of µ = 17, σ = 10. Say x̄100 is the mean of a sample of size 100
selected from the same sample.
1. What is the sampling distribution of x̄25?
Normal,
µx̄25= 17, σx̄25
= 2
2. What is the sampling distribution of x̄100?
Normal,
µx̄25= 17, σx̄100
= 1
3. Which of the probabilities, P(15 < x̄25 < 19) or
P(15 < x̄100 < 19), would you expect to be larger?
The latter.
4. Calculate the probabilities in the third question.
At z = 1,
area=0.3413, z = 2, area=0.4772. Thus the first is 0.6826, the
second is 0.9544.
Session 4: Samples and sampling distributions – p. 30
![Page 32: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/32.jpg)
Testing concepts: Comparing sam-pling distributions
Say x̄25 is the mean of a random sample of size 25 from a population
of µ = 17, σ = 10. Say x̄100 is the mean of a sample of size 100
selected from the same sample.
1. What is the sampling distribution of x̄25? Normal,
µx̄25= 17, σx̄25
= 2
2. What is the sampling distribution of x̄100? Normal,
µx̄25= 17, σx̄100
= 1
3. Which of the probabilities, P(15 < x̄25 < 19) or
P(15 < x̄100 < 19), would you expect to be larger? The latter.
4. Calculate the probabilities in the third question. At z = 1,
area=0.3413, z = 2, area=0.4772. Thus the first is 0.6826, the
second is 0.9544.Session 4: Samples and sampling distributions – p. 30
![Page 33: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/33.jpg)
Problems to be solved
Session 4: Samples and sampling distributions – p. 31
![Page 34: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/34.jpg)
Q1: Sampling distribution parame-ters
Suppose a random sample of n = 100 is selected from a population
with µ, σ as follows. Find the values of µx̄, σx̄
µ = 10, σ = 20
µ = 20, σ = 10
µ = 50, σ = 300
µ = 100, σ = 200
Session 4: Samples and sampling distributions – p. 32
![Page 35: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/35.jpg)
Q2: Sampling distribution probabili-ties
Suppose a random sample of n = 225 is selected from a population
with µ = 70, σ = 30. Find the following probabilities:
Pr(x̄ > 72.5)
Pr(x̄ < 73.5)
Pr(69.1 < x̄ < 74.0)
Pr(x̄ < 65.5)
Session 4: Samples and sampling distributions – p. 33
![Page 36: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/36.jpg)
Q3: Tobacco company research
Research by a tobacco company says that the relative freq.
distribution of the tar content of a new low-tar cigarette has µ = 3.9
mg of tar and σ = 1.0 mg. A sample of 100 low-tar cigs are selected
from one-day’s production and the tar content is measured:
What is Pr(mean tar content of the sample) is greater than
4.15mg?
Suppose that the sample mean tar content works out to be
x̄ = 4.18 mg. Based on the first question, do you think the
tobacco company may have understated the tar content?
If the tobacco company’s figures are correct, then rationalise
the observed value of x̄ = 4.18 mg.
Session 4: Samples and sampling distributions – p. 34
![Page 37: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/37.jpg)
Q4: Mean of an exponential RV
The length of time between arrivals at a hospital clinic and the
length of service are two RVs that are important in designing a
clinic, and how many doctors/nurses are needed there. Suppose the
relative freq. dist. of the interarrival time (between patients) has a
mean of 4.1 minutes and σ = 3.7 minutes.
1. A sample of 20 interarrival times are selected and x̄ the sample
mean is calculated. What is the sampling distirbution of x̄?
2. What is the Pr(the mean interarrival time) will be less than 2
minutes in this sample?
3. What is the Pr(the mean interarrival time) of the sample will
exceed 6.5 minutes?
4. Would you expect x̄ to be greater than 6.5 minutes? Explain.Session 4: Samples and sampling distributions – p. 35
![Page 38: Session 4: Samples and sampling distributionsPopulation Sample1 Sample2 Sample3 Mean 0.500 0.504 0.422 0.494 Sigma 0.289 0.294 0.203 0.289 1st Quartile 0.250 0.275 0.311 0.237 3rd](https://reader034.vdocuments.mx/reader034/viewer/2022042711/5f797b000b71737ae56e4f2a/html5/thumbnails/38.jpg)
References
Chapter 7,
Session 4: Samples and sampling distributions – p. 36