User:Tameem/Segment (9)

From Computational Statistics (CSE383M and CS395T)
Jump to navigation Jump to search


To Compute

1. Use characteristic functions to show that the sum of two independent Gaussian random variables is itself a Gaussian random variable. What is its mean and variance?

Doing it the hard way (using Mathematica):

Let X and Y be two independent Gaussian random variables, X has a mean <math>{\mu}_1</math>, and standard deviation <math>{\sigma}_1</math>, and Y has a mean <math>{\mu}_2</math>, and standard deviation <math>{\sigma}_2</math>. Let S = X + Y.

<math> p_S(s) = \int_{-\infty}^{\infty} p_X(u)p_Y(s-u) du = -\frac{1}{2\pi{\sigma}_1{\sigma}_2}\int_{-\infty}^{\infty} e^{\frac{1}{2}((\frac{u-{\mu}_1}{{\sigma}_1})^2-(\frac{s-u-{\mu}_2}{{\sigma}_2})^2)} du</math>

Using Mathematica:

Seg9 1.png

Clearly, it is a Guassian Distribution with mean = <math>{\mu}_1+{\mu}_2</math>, and variance <math>{{\sigma}_1}^2+{{\sigma}_2}^2</math>

2. Calculate (don't just look up) the characteristic function of the Exponential distribution.

<math>\Phi_X(t)= \int_0^{\infty} \beta e^{-(\beta-it)x} dx = \frac{\beta}{-(\beta-it)} e^{-(\beta-it)x}|_0^\infty = \frac{\beta}{-(\beta-it)}[0-1] = \frac{\beta}{(\beta-it)} </math>

Class Activity

Team 2, Teamed with with User:Noah, User:Kai, Keerthana Kumar's Solution, User:Jzhang, User:Trettels.

You can refer to: Kai for class activities.