Problem Set 2 in Economics 671: Multivariate and Conditional Probability, Assignments of Introduction to Econometrics

Five economics problems related to multivariate and conditional probability. Topics include valid bivariate density functions, marginal densities, independence of random variables, and conditional probability. Students are asked to derive answers using given formulas and provide justifications.

Typology: Assignments

Pre 2010

Uploaded on 09/02/2009

koofers-user-9ck-1
koofers-user-9ck-1 🇺🇸

3

(2)

10 documents

1 / 2

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Economics 671
Problem Set #2
Multivariate and Conditional Probability
(1) Consider the joint density function
f(x, y) = ½2 exp([x+y]) for 0 xy , 0y
0 otherwise .
(a) Show that this is a valid bivariate density function.
(b) Derive the marginal densities for Xand Y.
(c) Are the random variables Xand Yindependent?
(2) Suppose that Corr(X, Y ) = 0. Does this imply that the random variables Xand Yare also
independent? (Provide a proof or a counter example).
(3) Suppose that an individual is diagnosed with a disease x. This individual then goes on the
internet to “research” the disease and finds, in repeated searches, that a far more serious disease
yis associated with x. That is, the individual finds (and let’s treat this as a medical certainty)
that 80 percent of the individuals who have disease yalso have the (less serious) disease x. Our
individual then becomes quite concerned that she must also have disease y.
Is this necessarily the case? (Justify your answer in terms of conditional probability. If you believe
the concerns are justified, prove it; it you think they are unjustified provide a convincing counter-
example).
pf2

Partial preview of the text

Download Problem Set 2 in Economics 671: Multivariate and Conditional Probability and more Assignments Introduction to Econometrics in PDF only on Docsity!

Economics 671 Problem Set # Multivariate and Conditional Probability

(1) Consider the joint density function

f (x, y) =

{ (^) 2 exp(−[x + y]) for 0 ≤ x ≤ y, 0 ≤ y 0 otherwise.

(a) Show that this is a valid bivariate density function.

(b) Derive the marginal densities for X and Y.

(c) Are the random variables X and Y independent?

(2) Suppose that Corr(X, Y ) = 0. Does this imply that the random variables X and Y are also independent? (Provide a proof or a counter example).

(3) Suppose that an individual is diagnosed with a disease x. This individual then goes on the internet to “research” the disease and finds, in repeated searches, that a far more serious disease y is associated with x. That is, the individual finds (and let’s treat this as a medical certainty) that 80 percent of the individuals who have disease y also have the (less serious) disease x. Our individual then becomes quite concerned that she must also have disease y.

Is this necessarily the case? (Justify your answer in terms of conditional probability. If you believe the concerns are justified, prove it; it you think they are unjustified provide a convincing counter- example).

(4) In a popular game show, a contestant is asked to pick one of three boxes. One of the three boxes is known to contain a prize, while the other two are empty. The game show host knows the location of the prize. After the contestant makes her initial selection, the game show host opens one of the two boxes which was not chosen (and was known by the host to be empty), and shows her that this box is empty. The contestant then has the option to either stay with her original choice, or switch to the other unopened box. If the contestant wishes to maximize her probability of getting the prize, what should she do? (Stay with her original choice, or switch?)

(5) Consider an arbitrary bivariate density function p(x, y). Evaluate whether the following state- ments are true or false in general:

(a) The marginal densities pX (x) and pY (y) are sufficient to define (that is, completely characterize) the joint density p(x, y).

(b) The densities pX (x) and pY |X (y|x) are sufficient to define the joint.

(c) The densities pX (x) and pX|Y (x|y) are sufficient to define the joint.

(d) The densities pX|Y (x|y) and and pY |X (y|x) are sufficient to define the joint.

(Some of these may be obvious, but one of these is certainly not).