How to put stoploss in backtesting

Quant Trading Strategies & Ideas
1

dear sir,



in this strategy ther is only buy and for exit it is double selling what to code for exit or stoploss for backtesting

 df['Signal'] = np.nan

    df.loc[(df['SAR']< df['Adj Close']) & (df['fastd']>df['slowd']) & (df['fastk']>df['slowk']),'Signal'] =1

    df.loc[(df['SAR']> df['Adj Close']) & (df['fastd']<df['slowd']) & (df['fastk']<df['slowk']),'Signal'] =-1

    df['Signal'] = df['Signal'].fillna(method='ffill')

how to put 0 for exit. please suggest with if conditon i have tried but error occurs

2

Hello Promod,



You can incorporate stop-loss into your signal labels by perhaps adding more conditions which check for prices. If the adjusted close is lower than a threshold at any given point in the series you can make the signal as -1. 



df.loc[threshold> df['Adj Close']  ,'Signal'] =-1



You can do the same thing using returns as well. Set a percentage threshold and then check if returns are lower than this threshold. 



df.loc[threshold_ret> df['returns']  ,'Signal'] =-1



Similarly, you can put conditions for when you don't want to take a position using something like:



df.loc[condition  ,'Signal'] = 0



Do get back if further clarification is needed. 







 

3

thank you sir for your responce i have tried this but still getting error in code, the code is as follows, please suggest me where i am wrong what chages to be made.

 df['Signal'] = np.nan

    df.loc[(df['SAR']< df['Adj Close']) & (df['fastd']>df['slowd']) & (df['fastk']>df['slowk']),'Signal'] =1

    df.loc[(df['SAR']> df['Adj Close']) & (df['fastd']<df['slowd']) & (df['fastk']<df['slowk']),'Signal'] =-1

    df['Signal'] = df['Signal'].fillna(method='ffill')    

        if df.loc[(df['Signal'] == 1:

            df.loc[(df['Adj Close'] < (df['Adj Close']-(df['ATR']*2))), 'Signal'] = 0

        elif df.loc[(df['Signal'] == -1:

            df.loc[(df['Adj Close'] > (df['Adj Close']+(df['ATR']*2)), 'Signal'] = 0

    df['Signal'] = df['Signal'].fillna(method='ffill')

 

4

Hello Promod,



There are syntactical errors in this code. I would recommend that you try out Quantra's free offerings on pandas and NumPy. 



Your thought process is however right. You can do the first one as follows:



df.loc[(df['Signal'] == 1) & (df['Adj Close'] < (df['Adj Close']-(df['ATR']*2))), 'Signal'] = 0



No need for the if and elif.