accessnash’s gists

accessnash / stratsampling.M

Last active June 11, 2018 19:20

Stratified sampling example for a call option

	#Model Parameters
	S0 = 100;
	K =90;
	r = 0.025;
	sigma = 0.2;
	T = 1;

	#Simulation parameters
	#B = [1, 2, 5, 10, 20, 50, 100, 200, 500, 1000]; #Intervals
	#NB = [1000, 500, 200, 100, 50, 20, 10, 5, 2, 1]; # sample size

accessnash / asian.M

Last active June 12, 2018 03:53

Asian Options : Comparison of the Variance Ratios using 3 methods : Importance Sampling using GHS method to find µ , Stratification based on conditioning using the same µ vector, Combination of the Importance Sampling and Stratification

	r = .05; % Set up parameters and arrays
	sig = .10;
	T = 1;
	s0 = 50;
	K = 45;
	N = 16; %time steps along path
	n = 1000000;
	B=100; % number of bins for stratification
	NB = n/B; %sample size per bin
	del = T/N;

accessnash / class-size

Created June 13, 2018 01:44

Maimonides rules - Stata

	reg avgmath classize
	reg avgverb classize

	reg avgmath classize ses
	reg avgverb classize ses

	gen maimrule = gradesize/(int((gradesize-1)/40)+1)
	twoway (scatter classize gradesize, msize(small)) (line maimrule gradesize, lcolor(orange))

	reg avgmath maimrule gradesize

accessnash / splitpersonality.py

Created June 13, 2018 23:37

A function to split names and keep only title and last names

	Personalities = ['Mr. Clarke Kent', 'Dr. Victor Doom', 'Dr. Stephen Strange', 'Dr. Jonathan Osterman']

	def split_names(superhero):
	title = superhero.split()[0]
	lastname = superhero.split()[-1]
	return '{} {}'.format(title, lastname)

	list(map(split_names, Personalities))

accessnash / pca_stockreturns.R

Created June 17, 2018 00:45

Principal component analysis and factor analysis for stock return data

	findata <- read.table("C:/Users/DASA0/Desktop/MS Econ/Stat 524/wichern data/T8-4.dat", sep="\t", header=F)
	colnames(findata) <- c("JP Morgan", "City bank", "Wells Fargo", "Royal Dutch", "Exxon")
	Xbar <- colMeans(findata)
	S <- cov(findata)
	weight <- diag(1/sqrt(diag(cov(findata))))
	#standardization
	stdfindata <- (as.matrix(findata)-rep(1, dim(findata)[1])%%t(apply(findata, 2, mean)))%%weight
	stdfindata <- as.data.frame(stdfindata)
	colnames(stdfindata) <- c("JP Morgan", "City bank", "Wells Fargo", "Royal Dutch", "Exxon")
	R <- cov(stdfindata)

accessnash / balancedbrackets.py

Created June 24, 2018 00:59

Balanced brackets problem from Hackerrank

	def balanced( a_str):
	if len(a_str) %2 !=0:
	return False

	open_bracks = ['(', '{', '[']
	close_bracks = [')', '}', ']']
	par_dict = {open_bracks[0]:close_bracks[0], open_bracks[1]:close_bracks[1], open_bracks[2]:close_bracks[2]}

	if a_str[0] in close_bracks:
	return False

accessnash / melonsboxes.py

Created June 24, 2018 01:01

Melons and boxes problem from Hackerrank (need to revise - all test cases do not pass)

	def melon_count(boxes, melons):
	boxes.sort()
	melons.sort()
	b= len(boxes)
	m = len(melons)

	noofmelons = 0
	i = 0
	j = 0
	while(i != b and j != m):

accessnash / dfexample.py

Last active July 3, 2018 21:06

A simple example of Pandas and Dataframes

	import pandas as pd

	purchase_1 = pd.Series({'Name': 'Paul',
	'Book Purchased': 'Monte Carlo Methods',
	'Cost': 94.69})
	purchase_2 = pd.Series({'Name': 'Steve',
	'Book Purchased': 'Stochastic Calculus',
	'Cost': 39.55})
	purchase_3 = pd.Series({'Name': 'Brigo',
	'Book Purchased': 'Interest Rate Models',

accessnash / binarysearch.py

Last active July 9, 2018 20:50

Binary search algorithm with a measurement of performance in terms of time taken

	from time import time

	def present(collection, num):
	'''finds out if number of our interest is present in collection'''
	return num in collection

	def binsearch(ordered, num):
	'''binary search function to show position of number of our interest in the collection'''
	low = 0
	high = len(ordered)-1

accessnash / chisquare.py

Last active July 7, 2018 03:31

Checking how skew in a chi-squared distribution decreases as the degrees of freedom increases

	import numpy as np
	import pandas as pd
	import scipy.stats as stat
	import matplotlib.pyplot as plt

	chi_squared_df1 = np.random.chisquare(1, size = 100000)

	chi_squared_df2 = np.random.chisquare(2, size = 100000)

	chi_squared_df3 = np.random.chisquare(3, size = 100000)