import numpy as np
import matplotlib.pyplot as plt
import bivariate
import math
carat = np.loadtxt('diamond.tab', delimiter='\t', skiprows=1,usecols=[0])
price = np.loadtxt('diamond.tab', delimiter='\t', skiprows=1,usecols=[1])
data = zip(carat, price) #sort
data.sort()
(carat, price) = zip(* data) # unzip
bivariate.bivariate(carat,price)
plt.plot(carat, price, 'k+')
plt.xlabel("Carats")
plt.ylabel("Price")
plt.title("Price of Diamonds by Carat")
plt.show()
print beta1, alpha1, beta2, alpha2
allYears=[y for y in pre98Years]
allYears.extend(post98Years)
print allYears
alltemps=[t for t in pre98temps]
alltemps.extend(post98temps)
crossyhat1=-np.array([alpha1+beta1* y for y in allYears]) +alltemps
crossyhat2=-np.array([alpha2+beta2* y for y in allYears]) +alltemps
print "A",crossyhat1, np.sum(crossyhat1)
print "B",crossyhat2, np.sum(crossyhat2)
print "C", crossyhat2-crossyhat1, sum(crossyhat2-crossyhat1)
beta3,alpha3=regress.regress(np.array(alltemps),np.array(allYears))
crossyhat3=-np.array([alpha3+beta3* y for y in allYears]) +alltemps
for i in range(len(crossyhat1)):
print allYears[i], alltemps[i], crossyhat1[i], crossyhat2[i], allYears[i]*beta1+alpha1,allYears[i]*beta2+alpha2, crossyhat3[i],allYears[i]*beta3+alpha3
import bivariate
import random
bv=bivariate.bivariate(crossyhat1,np.array([random.random() for c in crossyhat1]), anomalise=False, pr=0.01)
bv2=bivariate.bivariate(crossyhat2,np.array([random.random() for c in crossyhat2]), anomalise=False, pr=0.01)
bv3=bivariate.bivariate(crossyhat3,np.array([random.random() for c in crossyhat3]), anomalise=False, pr=0.01)
bv4=bivariate.bivariate(alltemps,np.array([random.random() for c in alltemps]), anomalise=False, pr=0.01)
# -*- coding: utf-8 -*- """ Created on Mon Aug 25 17:12:14 2014 @author: s4493222 """ SVNRevision="$Revision: 307 $" #stability.py - code for examining the stability of breakpoints, especially near the start of finish of data import bivariate import tests17Aug import numpy as np import random #def stability(testData, controlData, yearData): data = np.genfromtxt(tests17Aug.fn,delimiter=",",names=True,filling_values =np.NaN) ys=data["B4"] l=len(ys) std=np.std(ys) brk=l/2+20 ys[brk:]+=std/2 seg=brk-20 Years=data["Year"] xs=np.array([random.random() for y in Years]) for seg in range(brk+10): bv=bivariate.bivariate(ys[seg:l-34], xs[seg:l-34], anomalise=False, pr=0.01) print seg, Years[seg+1], Years[brk+1], bv.stepChange(), Years[bv.maxIndexTi()+1+seg], std, bv.maxTi(),bivariate.Pr2(bv.maxTi(),99), bv.critical()
import numpy as np
import matplotlib.pyplot as plt
import bivariate
import math
carat = np.loadtxt('diamond.tab', delimiter='\t', skiprows=1, usecols=[0])
price = np.loadtxt('diamond.tab', delimiter='\t', skiprows=1, usecols=[1])
data = zip(carat, price) #sort
data.sort()
(carat, price) = zip(*data) # unzip
bivariate.bivariate(carat, price)
plt.plot(carat, price, 'k+')
plt.xlabel("Carats")
plt.ylabel("Price")
plt.title("Price of Diamonds by Carat")
plt.show()
