-
Notifications
You must be signed in to change notification settings - Fork 29
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
added stats/Fisher-information-derivation.ipynb
- Loading branch information
Zhuyi Xue
committed
Apr 13, 2022
1 parent
c3798fa
commit da33881
Showing
1 changed file
with
120 additions
and
0 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,120 @@ | ||
{ | ||
"cells": [ | ||
{ | ||
"cell_type": "markdown", | ||
"metadata": {}, | ||
"source": [ | ||
"Rederives https://en.wikipedia.org/wiki/Fisher_information#Definition." | ||
] | ||
}, | ||
{ | ||
"cell_type": "markdown", | ||
"metadata": {}, | ||
"source": [ | ||
"given $f(X; \\theta)$ is the maximum likelihood of $\\theta$ based on data $X$," | ||
] | ||
}, | ||
{ | ||
"cell_type": "markdown", | ||
"metadata": {}, | ||
"source": [ | ||
"\\begin{align*}\n", | ||
"\\frac{\\partial}{\\partial \\theta} \\log f(X; \\theta) \n", | ||
"&= \\frac{\\frac{\\partial}{\\partial \\theta} f(X; \\theta)}{ f(X;\\theta)} \\\\\n", | ||
"\\frac{\\partial^2}{\\partial \\theta^2} \\log f(X; \\theta) \n", | ||
"&= \\frac{\\left( \\frac{\\partial^2}{\\partial \\theta^2} f(X; \\theta) \\right ) f(X;\\theta) - \\frac{\\partial}{\\partial \\theta} f(X; \\theta) \\frac{\\partial}{\\partial \\theta} f(X ;\\theta) }{ f(X;\\theta) ^2} \\\\\n", | ||
"&= \\frac{\\frac{\\partial^2}{\\partial \\theta^2} f(X; \\theta)}{ f(X;\\theta)} - \\left( \\frac{\\frac{\\partial} {\\partial \\theta} f(X; \\theta)}{f(X ;\\theta)} \\right )^2 \\\\\n", | ||
"&= \\frac{\\frac{\\partial^2}{\\partial \\theta^2} f(X; \\theta)}{ f(X;\\theta)} - \\left( \\frac{\\partial}{\\partial \\theta} \\log f(X; \\theta) \\right )^2\n", | ||
"\\end{align*}" | ||
] | ||
}, | ||
{ | ||
"cell_type": "markdown", | ||
"metadata": {}, | ||
"source": [ | ||
"The expectation of the first part is equal to zero" | ||
] | ||
}, | ||
{ | ||
"cell_type": "markdown", | ||
"metadata": {}, | ||
"source": [ | ||
"\\begin{align*}\n", | ||
"\\mathbb{E} \\left[ \\frac{\\frac{\\partial^2}{\\partial \\theta^2} f(X; \\theta)}{ f(X;\\theta)} \\bigg| \\theta \\right ]\n", | ||
"&= \\int_{X} \\frac{\\frac{\\partial^2}{\\partial \\theta^2} f(X; \\theta)}{ f(X;\\theta)} f(X; \\theta) dX \\\\\n", | ||
"&= \\int_{X} \\frac{\\partial^2}{\\partial \\theta^2} f(X; \\theta) dX \\\\\n", | ||
"&= \\frac{\\partial^2}{\\partial \\theta^2} \\int_X f(X;\\theta) dX \\\\\n", | ||
"&= \\frac{\\partial^2}{\\partial \\theta^2} 1 \\\\\n", | ||
"&= 0\n", | ||
"\\end{align*}" | ||
] | ||
}, | ||
{ | ||
"cell_type": "markdown", | ||
"metadata": {}, | ||
"source": [ | ||
"So" | ||
] | ||
}, | ||
{ | ||
"cell_type": "markdown", | ||
"metadata": {}, | ||
"source": [ | ||
"\\begin{align*}\n", | ||
"\\mathbb{E}\\left[ \\frac{\\partial}{\\partial \\theta} \\log L(X; \\theta) \\bigg | \\theta \\right ]\n", | ||
"&= - \\mathbb{E}\\left[ \\left( \\frac{\\partial}{\\partial \\theta} \\log f(X; \\theta) \\right )^2 \\bigg | \\theta \\right]\n", | ||
"\\end{align*}" | ||
] | ||
}, | ||
{ | ||
"cell_type": "markdown", | ||
"metadata": {}, | ||
"source": [ | ||
"which is the negative of variance of the score $\\frac{\\partial}{\\partial \\theta} \\log f(X; \\theta)$ since the expectation of the score is also 0:" | ||
] | ||
}, | ||
{ | ||
"cell_type": "markdown", | ||
"metadata": {}, | ||
"source": [ | ||
"\\begin{align*}\n", | ||
"\\mathbb{E}\\left[ \\frac{\\partial}{\\partial \\theta} \\log f(X; \\theta) | \\theta \\right] \n", | ||
"&= \\mathbb{E}\\left[ \\frac{ \\frac{\\partial}{\\partial \\theta} f(X; \\theta)}{f(X;\\theta)} \\bigg| \\theta \\right] \n", | ||
"&= \\int_X \\frac{ \\frac{\\partial}{\\partial \\theta} f(X; \\theta)}{f(X;\\theta)} f(X;\\theta) dX \\\\\n", | ||
"&= \\int_X \\frac{\\partial}{\\partial \\theta} f(X; \\theta) dX \\\\\n", | ||
"&= \\frac{\\partial}{\\partial \\theta} \\int_X f(X; \\theta) dX \\\\\n", | ||
"&= \\frac{\\partial}{\\partial \\theta} 1 \\\\\n", | ||
"&= 0\n", | ||
"\\end{align*}" | ||
] | ||
}, | ||
{ | ||
"cell_type": "code", | ||
"execution_count": null, | ||
"metadata": {}, | ||
"outputs": [], | ||
"source": [] | ||
} | ||
], | ||
"metadata": { | ||
"kernelspec": { | ||
"display_name": "Python 3", | ||
"language": "python", | ||
"name": "python3" | ||
}, | ||
"language_info": { | ||
"codemirror_mode": { | ||
"name": "ipython", | ||
"version": 3 | ||
}, | ||
"file_extension": ".py", | ||
"mimetype": "text/x-python", | ||
"name": "python", | ||
"nbconvert_exporter": "python", | ||
"pygments_lexer": "ipython3", | ||
"version": "3.8.6" | ||
} | ||
}, | ||
"nbformat": 4, | ||
"nbformat_minor": 4 | ||
} |