Data Analysis Yahoo Finance Data

What the Script does :
=> Downloads historical Stock Prices for a company from Yahoo Finance website.
=> Calculates the 10 day moving average.
=> Generate the graph for the data


a) Necessary Imports:
1. Import Panda. Panda is a data analysis library of python.  Pandas has tool to read and write between in-memory data structures and different file formats. It has efficient data frame object for data manipulation with better indexing support.
2. Import Datetime: Provides classes to manipulate date and time objects. Api to convert between different file formats.
3. Import matplotlib: Python 2D plotting library , simple to use and genrates graphs, plots etc with few lines of code. 
4. Import Numpy : Scientific computing package for python,N-Dimensional array object ,Linear algebra related functions.
5. Import urlib: URL handling module for python

import pandas as pd

import pandas.io.data

from datetime import timedelta

import datetime as dt

from pandas import Series, DataFrame

import matplotlib.pyplot as plt

import matplotlib as mpl

import urllib.request

import numpy as numpy

from datetime import datetime

from matplotlib.pyplot import *

import matplotlib.dates as mdates

b) Create a class called Stock 
The main(Init) function should accept as parameter the company symbol, lookback period, window size,and the end date.

def __init__(self,symbol,lookback_period,window_size,end=datetime.date.today()): 

Suppose your lookback period is 100 then get prices for 100 days.
SO to get prices for 100 days subtract from the end date the lookback period.
But for that you need to use same format . 
In this example I have used timedelta.  Time delta helps you get the start date by specifying the number of days  from given date. Like your end date is today and you want stoc prices for last 100 days ,we can use timedelta.

start = end - timedelta(days=lookback_period)

c) Convert date into required format.

start_date = start.isoformat() 

d)  get the required Url for data analysis. 

url = "http://ichart.finance.yahoo.com/table.csv?s={0}".format(symbol)
url += "&a={0}&b={1}&c={2}".format(start_month,start_day,start_year)
url += "&d={0}&e={1}&f={2}".format(end_month,end_day,end_year)

e) Parse data

df = pd.read_csv(urllib.request.urlopen(url))
#get the adj close from csv
saved_column = df['Adj Close']
#get the matching date
y_data = df['Date']

f) Get the moving average

def movingaverage(self,interval, window_size):
window = numpy.ones(int(window_size))/float(window_size)
return numpy.convolve(interval, window, 'same')

Moving Average smooths price fluctuations by removing the noise. It computes the averages of a subset of full data set.

Moving Average Wiki 

g) Generate the graph

 x_data = x_points[0:70]
#get the moving average
y_av = self.movingaverage(saved_column,window_size)
#generate graph
figure("Plot of stocks")
x = [dt.datetime.strptime(d,"%Y-%m-%d").date() for d in x_data]
gca().xaxis.set_major_formatter(mdates.DateFormatter("%Y-%m-%d"))
gca().xaxis.set_major_locator(mdates.DayLocator())
plot(x,y_av,'r')
plot(x,close,'g')
gcf().autofmt_xdate()
xlabel("Date")
ylabel("adjusted close")
show()
GIT