Scribd
Upload
38 views

Final Practice Problems

This document contains 4 practice problems related to probability and statistics: 1) A crime scenario where the probability of guilt given evidence is calculated based on prior probabilities. 2) Bayesian analysis of geometric data with discrete and beta priors, including posterior distributions and estimation. 3) Identification of which time series processes are stationary, and specifying their means and autocovariance functions. 4) Markov chain analysis for a weather model, including initial state probabilities, transitional probabilities over time, and MLE estimation from data.

Uploaded by

Kanakamedala Sai Rithvik ee18b051
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
38 views

Final Practice Problems

This document contains 4 practice problems related to probability and statistics: 1) A crime scenario where the probability of guilt given evidence is calculated based on prior probabilities. 2) Bayesian analysis of geometric data with discrete and beta priors, including posterior distributions and estimation. 3) Identification of which time series processes are stationary, and specifying their means and autocovariance functions. 4) Markov chain analysis for a weather model, including initial state probabilities, transitional probabilities over time, and MLE estimation from data.

Uploaded by

Kanakamedala Sai Rithvik ee18b051
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

GR5703

Spring 2023 — Section 2


Practice Problems — Final

1. A crime has been committed. Assume that the crime was committed by exactly one
person, and that there are 1000 people who could have committed the crime and that,
in the absence of any evidence, these people are all equally likely to be guilty of the
crime. Suppose a piece of evidence is found and a particular individual is accused. It
is judged that this evidence would have a probability of 0.99 of being observed if the
crime were committed by that individual, and a probability of only 0.0001 of being
observed, if the crime were committed by any other individual. Find the probability,
given the evidence, that the individual committed the crime.

2. Consider a Bayesian model for geometric data X ∼ Geo(q). That is,

f (x|θ) = θ(1 − θ)x−1 , x = 1, 2, . . .


π(θ) =?, θ ∈ [0, 1]

(a) Suppose there is only one observation x = 13, and the prior is discrete uniform
on {0, 1/6, 2/6, 3/6, 4/6, 5/6, 1}. Derive the posterior probability mass f (θ|x) and
calculate the posterior mean as an estimator of θ.
(b) Now consider a random sample X1 , . . . , Xn and Beta(a, b) prior for θ, that is,
1
π(θ) = θα−1 (1 − θ)β−1
B(α, β)

Derive the posterior pdf f (θ|x1 , . . . , xn ) and use it to show that the Beta distri-
bution is a conjugate prior for this model.
(c) In the setup of part (b), suppose the sample is 20, 1, 2, 14, 1, 5, 2, 3, 9, 2 and
that the parameters of the Beta prior are α = 6, β = 2. Graph the posterior pdf
and estimate θ with the posterior mean.

3. Let {Zt } be a sequence of independent normal random variables, each with mean 0
and variance σ 2 , and let a, b, and c be constants. Which, if any, of the following
processes are (weakly) stationary? For each stationary process specify the mean and
the autocovariance function.

(a) Xt = Z1 cos(ct) + Z2 sin(ct)


(b) Xt = Z1 cos(ct) + Zt−1 sin(ct)
(c) Xt = a + bZ0

1
4. Suppose the following transition matrix was proposed for a Markov chain {Xn } that
models the transition between rainy and non-rainy days (where the first row / column
represents rain and the second row / column represents non-rain):
 
0.8 0.2
0.5 0.5

Suppose further the initial distribution of X0 is (0.5, 0.5).

(a) Find the probability that there is rain on Day 1


(b) Find the probability that there is no rain on Day 3, given that it was raining on
Day 0
(c) In the long run, what is the proportion of rainy days?
(d) Suppose the following data were collected over 3 years: π11 = 362, π12 = 186, π21 =
136, π22 = 157 where 1 = rain and 2 = no rain and πij stands for the observed
frequency of transitions from i to j, i, j ∈ {1, 2}. Calculate the MLE of the
transition matrix P.

You might also like