Getting data from Google Finance

In this lecture we’ll create a python code that can be used to download historical prices from Google Finance.

The task is a simple one, we just need to find the symbol of the stock and the exchange, and our program will create the appropiate URL ;).

Note

Seems that you can’t download historical data for currencies, this method works only for equity instruments.

Now let’s get started!

Requirements

For you to follow this lecture please make sure that you have the following python package installed:

pandas (I used version 0.18.0 for this lecture).

Explaining the URL (The theory)

The URL that we’ll need, looks something like this:

http://finance.google.com/finance/historical?q=BMV:AMXL&startdate=Jan+3+2000&enddate=Nov+11+2017&output=csv

At first, this might be daunting but the only parts you have to worry about are:

q=BMV:AMXL this part makes reference to the exchange where our stock is listed (in this case Bolsa Mexicana de Valores - Mexican Stock Exchange) and the symbol for identifying the company (in this case America Movil Serie L)
startdate=Jan+3+2000
enddate=Nov+11+2017

I think there is no need for me to explain you the last two parameters you’re smart enough to figure out what they are ;).

The remaining parts of the URL are kept as they are, so basically our code only needs the above three parameters and no more.

And finally the code (The practice)

import pandas as pd

def getGooglePrices(exchanges,symbols,startDate,endDate):
    '''
    Gets historical data from Google Finance
    Parameters:
    
    exchanges = list of the exchanges according to the nomenclature
    used by Google Finance
    
    symbols = list of stock symbols according to the symbols from
    Google Finance (Same length) as exchanges
    
    The correspondance must have the form Ex1:Sym1, Ex2:Sym2,...etc
    
    startDate = Start date in format Mmm+dd+yyyy
    
    endDate = End date in format Mmm+dd+yyyy

    Output
    
    No output, the data is stored in the current working directory
    as CSV file.
    
    '''
    #Declare the static parts of the URL
    urlStatic1="http://finance.google.com/finance/historical?q="
    urlStatic2="&output=csv"

    #Create the dynamic parts of the url and downloads a file
    #per symbol
    
    for i in range(0,len(exchanges)):
        quote=exchanges[i] + ":" + symbols[i]
        url=urlStatic1 + quote + "&startdate=" + startDate\
        + "&enddate=" + endDate + urlStatic2
            
        #Read the data from the url
        data=pd.read_csv(url)
            
        #Save it in the current directory
        #Removes the column index
        fileName=exchanges[i] + "_" + symbols[i] + ".csv"
        data.to_csv(fileName,index=False)

Testing the code

Now let’s test our code with the following examples

exchanges=["BMV","BMV","NYSEARCA"]
symbols=["AMXL","KOFL","SPY"]
startDate="Jan+01+2017"
endDate="Nov+10+2017"
getGooglePrices(exchanges,symbols,startDate,endDate)

The above lines should have created 3 CSV files each containing the historical prices from January 3rd 2017 (01 and 02 were not working days) up to and including November 10th 2017.

Try with your own symbols.

Conclusion

As you can see, the code has not error handlers so make sure that the symbols and exchanges you enter are the correct ones.

Also you can implement a little function so you can read these variables from a csv file avoiding you having to type them manually, but I won’t take all the fun from you :)

Hope you find the above useful

Thanks for reading!

Home Page

Getting data from Google Finance

David Montalván