Scribd
Upload
74 views5 pages

Advanced Statistics Assignment Help

The document discusses using a multinomial distribution to model gene frequency data from a population sample of 50 people. It provides the following information: 1) The maximum likelihood estimate (MLE) of the parameter θ can be found by differentiating the log likelihood function. 2) The asymptotic variance of the MLE is based on the second derivative of the log likelihood and properties of the binomial distribution. 3) The asymptotic distribution of the MLE is normal. 4) A 90% confidence interval for θ can be constructed using the normal approximation and quantiles from a standard normal distribution.

Uploaded by

Statistics Homework Solver
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
74 views5 pages

Advanced Statistics Assignment Help

The document discusses using a multinomial distribution to model gene frequency data from a population sample of 50 people. It provides the following information: 1) The maximum likelihood estimate (MLE) of the parameter θ can be found by differentiating the log likelihood function. 2) The asymptotic variance of the MLE is based on the second derivative of the log likelihood and properties of the binomial distribution. 3) The asymptotic distribution of the MLE is normal. 4) A 90% confidence interval for θ can be constructed using the normal approximation and quantiles from a standard normal distribution.

Uploaded by

Statistics Homework Solver
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 5

Advanced Statistics Assignment Help

For any Assignment related queries, call us at- +1 678 648 4277
You can mail us at:-
Hardy-Weinberg [email protected]
(Multinomial) Model of Gene Frequenciesor
reach us at- https://www.statisticshomeworksolver.com/

For a certain population, gene frequencies are in equilibrium: the


genotypes AA, Aa, and aa occur with probabilities (1 − θ)2, 2θ(1
− θ), and θ2. A random sample of 50 people from the population
yielded the following data:

The table counts can be modeled as the multinomial


distribution:
(X1, X2, X3) ∼ Multinomial(n = 50, p = ((1 − θ)2, 2θ(1 −
θ), θ2).
(a). Find the mle of θ
(b). Find the asymptotic variance of the mle. (c). What is the
large sample asymptotic distribution of the mle?
(d). Find an approximate 90% confidence interval for θ. To
construct the interval you may use the follow table of
cumulative probabilities for a standard normal N(0, 1) random
variable Z
(e). Using the mle θ ˆ in (a), 1000 samples from the

Multinomial(n = 50, p = ((1 − θ ˆ)2, 2θ ˆ(1 −


θ ˆ), θˆ2))
distribution were randomly generated, and mle estimates were
computed for each sample: θ ˆ j∗, j = 1, . . . , 1000.

For the true parameter θ0, the sampling distribution of Δ = θ ˆ−


θ0 is approximated by that of Δ˜ = θ ˆ∗ − θ. ˆ The 50-th largest
value of Δ˜ was +0.065 and the 50-th smallest value was −0.067.

Use this information and the estimate in (a) to construct a


(parametric) bootstrap confidence interval for the true θ0. What
is the confidence level of the interval? (If you do not have an
answer to part (a), assume the mle θ ˆ = 0.25).

Solution:
(a). Find the mle of θ

• (X1, X2, X3) ∼ Multinomial(n, p = ((1 − θ)2, 2θ(1 − θ), θ2))

• Log Likelihood for θ

ℓ(θ) = log(f(x1, x2, x3 | p1(θ), p2(θ), p3(θ)))

• First Differential of log likelihood:

(b). Find the asymptotic variance of the mle.


• Second Differential of log likelihood:

• Each of the Xi are Binomial(n, pi(θ)) so

E[X1] = np1(θ) = n(1 − θ)2


E[X2] = np2(θ) = n2θ(1 − θ)
E[X3] = np3(θ) = nθ2
(c) The asymptotic distribution of θ ˆ is

(d) An approximate 90% confidence interval


for θ is given by

So the approximate 90% confidence interval is:

{θ : 0.20 − .06580 < θ < 0.20 + .06580}

(e). For the bootstrap distribution of the errors Δ = θ ˆ− θ0,


(where θ0 is the true value), the approximate 5% and 95%
quantiles are

The approximate 90% confidence interval is

You might also like