import scipy

import matplotlib.pyplot as plt

#import statsmodels

import statsmodels.stats.stattools as stools

import scipy.optimize

import pandas as pd

filehandle=open('C:\\abs.xlsx','r')

print filehandle

'''standardization of NaOH'''

eb=0.1

ep=0.1

ewb=0.001

N_oxalic=0.2

V_oxalic=5

V_naoh=22.9

'''Normality of NaOH= (N_oxalic*V_oxalic/V_naoh)

Error(normality)/normality=((error(V_oxalic)/V_oxalic)^2-(error(V_naoh)/V_naoh)^2)^0.5

'''

e1=V_oxalic/ep

e2=V_naoh/eb

e3=(abs((e1**2.0)-(e2**2.0)))**0.5

N_naoh=N_oxalic*(V_oxalic/V_naoh)

e_standardization=e3*N_naoh

e=pd.ExcelFile('C:\\abs.xlsx')

sheet4=e.parse(3)

V_nacl=sheet4.icol(2).real

ep=sheet4.icol(8).real

V_mixture=sheet4.icol(10).real

eb=sheet4.icol(9).real

V_Naoh=sheet4.icol(3).real

N_hcl=N_naoh*V_Naoh/V_mixture

e4=V_mixture/ep

e5=V_Naoh/eb

e6=(abs((e4**2.0)-(e5**2.0)+(e3**2.0)))**0.5

e_strength=e6*N_hcl

#gm equivalents of HCl formed=x

x=(N_hcl/1000)*V_nacl

e7=V_nacl/ep

e8=(abs((e6**2.0)-(e7**2.0)))**0.5

e_x=e8*x

#(x/m) ratio

m=sheet4.icol(1).real

q=x/m

ewb=0.001

e9=(abs((e8**2.0)-(ewb**2.0)))**0.5

e_q=e9*q

M1=(x/m)+e_q

M2=(x/m)-e_q

c1=sheet4.icol(8).real

c=c1-N_hcl

e_c=(e_strength)*c

print "Normality of standardizeedd NaOH=:\n",N_naoh

print "strength of HCl in solution=:\n",N_hcl

print "gram equivalents of HCl:\n",x

print "value of x/m is:\n",q

print "error in calculation of the strength of HCl is:\n",e_strength

print "error in calculation of x:\n",e_x

print "error in calculation of x/m:\n",e_q

print "error in calculation of c:\n",e_c

print "absolute error in the experiment is:\n", e_q

u=1/(x/m)

t=1/c

def curve(t,p):

[A,B]=p

u=A+(B*t)

return u

def error(p,t,uexp):

ucalc=curve(t,p)

err=ucalc-uexp

return err

def get_r2(t,u,ucalc):

return r2,sigmai

umean=scipy.average(u)

sigmai=(((u-umean)**2)/9)**0.5

pguess=[1.050,0.587]

plsq=scipy.optimize.leastsq(error,pguess,args=(t,u))

p=plsq[0]

ucalc=curve(t,p)

j=u-ucalc

chisquare=sum((j**2)/((sigmai**2)*81))

sigmasq=sum(((u-ucalc)**2)/9)

sigmau=scipy.sqrt(sigmasq)

dumean2=(u-umean)**2

ducalc2=(u-ucalc)**2

dsducalc2=sum(ducalc2)

dsdumean2=sum(dumean2)

r2=1-dsducalc2/dsdumean2

residual=(u-ucalc)/e_q #calculating the residuals

w,p_shapiro=scipy.stats.shapiro(residual)#shapiro wilk test

dw=stools.durbin_watson(residual)#durbin watson test

DFM=8

DFE=1

squares=(ucalc-umean)

squaresT=(u-umean)

residuals=(ucalc-u)

SSM=sum(squares**2)

SSE=sum(residuals**2)

SST=sum(squaresT**2)

MSM=SSM/DFM

MSE=SSE/DFE

F=MSM/MSE

p_F=1-scipy.stats.f.cdf(F,DFM,DFE)

print "F test value is:",p_F

print "chisquare=:\n",chisquare

print "Durbin Watson parameter: \n",dw

print "r2=:\n",r2

print "Shapiro parameter:\n",w

print "sigmai=:\n",sigmai

print "uncertainty in data=:\n",sigmau

print "residuals are=:\n",residual

print "calculated values of log(x/m) from curve fit=:\n",ucalc

print "values of coefficients :\n", p

k=1/(x/m)

M1=ucalc+sigmau

M2=ucalc-sigmau

fig=plt.figure()

fig.show

axy=fig.add_subplot(111)

axy.title.set_text('m/x vs 1/c')

axy.plot(1/c,k,'o')

axy.plot(1/c,ucalc,label='data fit')

axy.plot(1/c,M1,label='Standard error of fit upper limit')

axy.plot(1/c,M2,label='Standard error of fit lower limit')

plt.legend()

fig=plt.figure()

fig.show

axy=fig.add_subplot(111)

axy.title.set_text('residuals vs u using fit')

axy.plot(u,residual,'*')