In: Statistics and Probability
Solution:
central limit theorem ;
for sample size,n>30
sample follows normal distribution
with mean =x bar
and stddev=sample stddev/sqrt(n)
a) Estimating a right skewed distribution like income
we cannot use central limit theorem for estimating right skewed dsitribution
(b)
b) Estimating the mean of a right skewed distribution like income with a large sample size
we can use central limit theorem
for n>30
which is large sample
for right skewed distribution mean >median
and mean is at the center line
c)Finding the exact probability of getting a proportion of successes less than a value
we can use central limit theorem
if population follows normal distribution or sample size is large
we can find
P(p^<p)
p^=sample proportion
p-population proportion
z=p^-p/sqrt(p*(1-p)/n
P(Z<p) can be found from standard normal tables.
and probability is nothing but area under normal curve.
d) Creating an approximate confidence interval for a proportion assuming normality.
we can use central limit theorem
assuming nomrality
95% confidence interval for true population proportion can be given as
p^-z*sqrt(p^(1-p^)/n,p^+z*sqrt(p^(1-p^)/n
z alpha/2 for 95%=1.96