CODEWITHSUNDEEP
pythonpandasfinance
MrProducer
MrProducer

Reputation: 3

Pandas and finance calculations

I am fairly new to using Python, and I am working on a stock analysis script. The idea is that the script will eventually take in a stock symbol, and the script will calculate Sharpe ratio, Treynor ratio, and other financial information. Right now, I am having trouble getting Pandas to work properly. I am unable to access just a column from the DataFrame to calculate yield for a stock.

from pandas.io.data import DataReader
from datetime import date, timedelta



def calc_yield(now, old):
    return (now-old)/old


def yield_array(cl):
    array = []
    count = 0
    for i in cl:
        old = cl[count]
        count += 1
        new = cl[count]
        array.append(calc_yield(new, old))
    return array


market = '^GSPC'
ticker = "AAPL"
days = 10

# set start and end dates
edate = date.today() - timedelta(days=1)
sdate = edate - timedelta(days=days)

# Read the stock price data from Yahoo
data = DataReader(ticker, 'yahoo', start=sdate, end=edate)

close = data['Adj Close']


print yield_array(close)

Error:

/Users/Tim/anaconda/bin/python "/Users/Tim/PycharmProjects/Test2/module tests.py"
Traceback (most recent call last):
  File "/Users/Tim/PycharmProjects/Test2/module tests.py", line 35, in <module>
    print yield_array(close)
  File "/Users/Tim/PycharmProjects/Test2/module tests.py", line 16, in yield_array
    new = cl[count]
  File "/Users/Tim/anaconda/lib/python2.7/site-packages/pandas/core/series.py", line 484, in __getitem__
    result = self.index.get_value(self, key)
  File "/Users/Tim/anaconda/lib/python2.7/site-packages/pandas/tseries/index.py", line 1243, in get_value
    return _maybe_box(self, Index.get_value(self, series, key), series, key)
  File "/Users/Tim/anaconda/lib/python2.7/site-packages/pandas/core/index.py", line 1202, in get_value
    return tslib.get_value_box(s, key)
  File "tslib.pyx", line 540, in pandas.tslib.get_value_box (pandas/tslib.c:11833)
  File "tslib.pyx", line 555, in pandas.tslib.get_value_box (pandas/tslib.c:11680)
IndexError: index out of bounds

Process finished with exit code 1

Upvotes: 0

Views: 1682

Answers (1)

TheBlackCat
TheBlackCat

Reputation: 10298

I think I see your problem. Given this function:

def yield_array(cl):
    array = []
    count = 0
    for i in cl:
        old = cl[count]
        count += 1
        print count
        new = cl[count]
        array.append(calc_yield(new, old))
        print old
        print new
    return array

The problem is that on the last item of cl, you will add 1 to count, which will result in an index one greater than the maximum index of cl. This results in an error, because it is trying to access an index that doesn't exist. You would need to do something like for i in cl[:-1], which would skip the last element.

However, there is a much simpler way to do this by vectorizing. You can reduce this whole function to:

close = data['Adj Close']
yield_data = close.diff()/close.shift(1)

or better yet, you can put the result back in the DataFrame for later use:

close = data['Adj Close']
data['Yield'] = close.diff()/close.shift(1)

Upvotes: 1

Related Questions