Question 4 Consider flipping a (biased) coin for which the probability of head is p. The fraction of heads after n independent tosses is Xn. Law of large numbers imply that Xn → p as n → o. This does not mean that Xn will exactly equal to p, but rather the distribution of X is tightly concentrated around p for large n. (a) Suppose 0.1 0.95? Note: The purpose of this question is to enhance your understanding of Chebyshev's inequality. In practice, the bound provided by Chebyshev's inequality is usually very "loose", in the sense that n actually only needs to be much smaller for P(0.5 < Xn < 0.7) > 0.95 to hold. A tighter bound for the setting we considered can be obtained by Hoeffding's inequality (Example 6.15 in Wasserman) or Central Limit Theorem (we will study this in class). Note that the bounds from Chebyshev's inequality or Hoeffding's inequality are finite sample in that they hold for any finite n. On the contrary, the bound from Central Limit Theorem is asymptotic, in that it is a statement about n o, and it only holds approximately for finite n.

Question 4 Consider flipping a (biased) coin for which the probability of head is p. The fraction of heads after n independent tosses is Xn. Law of large numbers imply that Xn → p as n → o. This does not mean that Xn will exactly equal to p, but rather the distribution of X is tightly concentrated around p for large n. (a) Suppose 0.1 0.95? Note: The purpose of this question is to enhance your understanding of Chebyshev's inequality. In practice, the bound provided by Chebyshev's inequality is usually very "loose", in the sense that n actually only needs to be much smaller for P(0.5 < Xn < 0.7) > 0.95 to hold. A tighter bound for the setting we considered can be obtained by Hoeffding's inequality (Example 6.15 in Wasserman) or Central Limit Theorem (we will study this in class). Note that the bounds from Chebyshev's inequality or Hoeffding's inequality are finite sample in that they hold for any finite n. On the contrary, the bound from Central Limit Theorem is asymptotic, in that it is a statement about n o, and it only holds approximately for finite n.

MATLAB: An Introduction with Applications

6th Edition

ISBN:9781119256830

Author:Amos Gilat

Publisher:Amos Gilat

Chapter1: Starting With Matlab

Section: Chapter Questions

Problem 1P

See similar textbooks

Similar questions

Consider the following loops on the Klein bottle. (1) Decide whether they are orientation reversing or preserving. Explain. (2) In each of the three case, cut the surface by the curve. What is the resulting surface for each case?
Differentiate
Prone then
Evealuate
help with these
Math 2: Unit 1 Review
A= A
Poove asb> a<
Calculation