In: Statistics and Probability
Taxes: The Internal Revenue Service reports that the mean federal income tax paid in the year
2010
was
$8040
. Assume that the standard deviation is
$4700
. The IRS plans to draw a sample of
1000
tax returns to study the effect of a new tax law.
(a) What is the probability that the sample mean tax is less than
$8000
? Round the answer to at least four decimal places.
The probability that the sample
mean tax is less than
$8000 is (b) What is the probability that the sample mean tax is between $7600 and$8100 ? Round the answer to at least four decimal places.
|
Solution:-
Given That:-
Let X be a random variable which represents the tax paid in 2010.
Given that, X ~ N(8040, 4700)
i.e. μ = $8040 and σ = $4700
We know that if X ~ N(μ , σ2) then x̄ ~ N(μ , σ2/n)
And if x̄ ~ N(μ , σ2/n) then (Dong z
(Where, x̄ is sample mean, n is sample size).
a) We have to obtain P(x̄ < $8000).
We have, μ = $8040, σ = $4700 and n = 1000
PT < 8000) = P (8000 - A) on / ov
P(i < 8000) = P(Z < (8000 – 8040) ) 4700/1000
Plī < 8000) = P(Z <-0.2691)
Using "pnorm" function of R we get, P(Z < -0.2691) = 0.3939
:. Plī < 8000) = 0.3939
Hence, the probability that the sample mean tax is less than $8000 is 0.3939.
b) We have to obtain P($7600 < x̄ < $8100).
We have, μ = $8040, σ = $4700 and n = 1000
\large P(7600<\bar{x}<8100)=P(\bar{x}<8100)-P(\bar{x}<7600)
\large P(7600<\bar{x}<8100)=P\left(\frac{(\bar{x}-\mu)}{\sigma/\sqrt{n}}<\frac{(8100-\mu)}{\sigma/\sqrt{n}} \right)- P\left(\frac{(\bar{x}-\mu)}{\sigma/\sqrt{n}}<\frac{(7600-\mu)}{\sigma/\sqrt{n}} \right)
\large P(7600<\bar{x}<8100)=P\left(Z<\frac{(8100-8040)}{4700/\sqrt{1000}} \right)- P\left(Z<\frac{(7600-8040)}{4700/\sqrt{1000}} \right)
\large P(7600<\bar{x}<8100)=P\left(Z<0.4037\right)- P\left(Z< -2.9604\right)Using "pnorm" function of R we get,
P(Z < 0.4037) = 0.6568 and P(Z < -2.9604) = 0.0015
\large \therefore P(7600<\bar{x}<8100)=0.6568-0.0015 = 0.6553
Hence, the probability that the sample mean tax is between $7600 and $8100 is 0.6553.
c) Let the 30th percentile of sample mean is k.
Hence, P(x̄ < k) = 0.30
\large P\left(\frac{(\bar{x}-\mu)}{\sigma/\sqrt{n}}<\frac{(k-\mu)}{\sigma/\sqrt{n}} \right)=0.30
\large P\left(Z<\frac{(k-8040)}{4700/\sqrt{1000}} \right)=0.30.....................(1)
Using "qnorm" function of R we get P(Z < -0.5244) = 0.30
Comparing, P(Z < -0.5244) = 0.30 and (1) we get,
\large \frac{(k-8040)}{4700/\sqrt{1000}} =-0.5244
\large k =8040+\left ( -0.5244\times\frac{4700}{\sqrt{1000}} \right )
he7962,06
Hence, the 30th percentile for the sample mean is
$7962.06.