Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
Reverse and inverse backtesting
Quant Trader
posted

hi,
This is just in my head for a couple of days, I am wondering if it would make sense to backtest an algo with the same data but reversed, that is from now back into the time, and to inverse the data by mirroring it, so a bull will become a bear and visaversa. Since 2009 the stock market has increased significantly...so maybe the data we use to backtest is too positive. At the moment we see a more side wards pattern. are the algos we make still viable? I see remarks in forums like "runs well till 2009, or since last years no returns.".

If others find it interesting to reverse and inverse data these might be nice features for Quantopian. Other idea is to have a random start. Maybe controllable byown code.

J..

1 response
David Edwards

J, testing on simulated data is always a good idea, and reversing real data is a simple way to get simulated prices. It makes sense, if the algorithm should make money regardless of market conditions, it should work with the data reversed. It can't be done in Quantopian, but zipline will run on whatever you throw at it, you could give that a shot.

Here's the generic zipline example with a function to reverse the index of the data it uses. 

import pytz  
import pandas as pd  
from datetime import datetime  
from zipline.utils.factory import load_from_yahoo  
from zipline import TradingAlgorithm  
from zipline.api import *


def initialize(context):  
    context.stock = 'AAPL'


def handle_date(context, data):  
    order(context.stock, 10)  

def reversed_data(data):  
    ''' Reverses the index of a pandas timeseries '''  
    type_ = type(data)  
    rdata = type_(data.values, index=data.index[-1::-1])  
    if type_ == pd.DataFrame:  
        rdata.columns = data.columns  
    return rdata.sort_index()


start = datetime(2005, 12, 22, 0, 0, 0, 0, pytz.utc)  
end = datetime(2012, 6, 3, 0, 0, 0, 0, pytz.utc)

data = load_from_yahoo(stocks=['AAPL'], indexes={}, start=start,  
                       end=end)  
data = data.dropna()  
data = reversed_data(data)  
algo = TradingAlgorithm(initialize=initialize,  
                        handle_data=handle_date)  
results = algo.run(data)  
results.portfolio_value.plot()
About Quantopian
Careers
Community
Events
Documentation
Academia
Lectures
Status
Facebook
Twitter
LinkedIn
YouTube
Terms of Use
Privacy Policy

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by Quantopian.

In addition, the material offers no opinion with respect to the suitability of any security or specific investment. No information contained herein should be regarded as a suggestion to engage in or refrain from any investment-related course of action as none of Quantopian nor any of its affiliates is undertaking to provide investment advice, act as an adviser to any plan or entity subject to the Employee Retirement Income Security Act of 1974, as amended, individual retirement account or individual retirement annuity, or give advice in a fiduciary capacity with respect to the materials presented herein. If you are an individual retirement or other investor, contact your financial advisor or other fiduciary unrelated to Quantopian about whether any given investment idea, strategy, product or service described herein may be appropriate for your circumstances. All investments involve risk, including loss of principal. Quantopian makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances.