Trading logic

[Forex2021]

As I am moving into vectorbt for backtesting I am trying to figure out the tradinglogic.
I get completely different results with the codes below. For some reason I get completely different results. I am not talking about the log returns, but a complete different equity curve. There is a shift in the signalvalues of 1 because of the trading delay of 1 day.
What could be the reason for such a big difference?

#####################
def backtest(frame,signals,settings):
try:

    prices=pd.DataFrame({})
    prices['date']=frame['date']
    prices['close']=frame['close']
    prices.set_index('date',inplace=True)
    
    rs = prices.apply(np.log).diff(1)
    
    prices['pos']=signals
    prices['pos']=prices['pos']
    prices['pos_shift1']=prices['pos'].shift(1)
    prices['rs']=rs
    
    prices['my_rs']=prices['pos_shift1']*prices['rs']
    prices['total']=prices['my_rs'].cumsum()
    #print (prices.head())
    prices['total']=pd.to_numeric(prices['total'])
    
    
    return prices

#################################

#ML model predict price
prediction=model.predict(prices)
signals = list(prediction.astype(int))
full_prices['signal']=signals
print ('prediction done')

full_prices['signal_shift']=full_prices['signal'].shift(1)

full_prices.set_index('date',inplace=True)

#vectorbt backtest
entries = full_prices['signal_shift']>0
exits = full_prices['signal_shift']<0
pf = vbt.Portfolio.from_signals(full_prices['close'], entries, exits, init_cash=100,size=1000,fees=0)
print (pf.stats())
pf.plot().show()

[polakowo]

Here's how to execute at the next open:

open = pd.Series([1, 2, 3, 4, 5])
close = pd.Series([2, 3, 4, 5, 6])
entries = pd.Series([False, True, False, False, False])
exits = pd.Series([False, False, False, True, False])

pf = vbt.Portfolio.from_signals(
    close,
    entries.vbt.signals.fshift(),
    exits.vbt.signals.fshift(),
    price=open
)
print(pf.orders.records_readable)
   Order Id  Column  Timestamp       Size  Price  Fees  Side
0         0       0          2  33.333333    3.0   0.0   Buy
1         1       0          4  33.333333    5.0   0.0  Sell

[Forex2021]

Thank you for the fast reply. I get the impression that this only takes long trades? I would need both directions...

…

-------- Original Message --------

On Jan 2, 2022, 23:05, Oleg Polakow wrote: Here's how to execute at the next open: open = pd . Series ([ 1 , 2 , 3 , 4 , 5 ]) close = pd . Series ([ 2 , 3 , 4 , 5 , 6 ]) entries = pd . Series ([ False , True , False , False , False ]) exits = pd . Series ([ False , False , False , True , False ]) pf = vbt . Portfolio . from_signals ( close , entries . vbt . signals . fshift (), exits . vbt . signals . fshift (), price = open ) print ( pf . orders . records_readable ) Order Id Column Timestamp Size Price Fees Side 0 0 0 2 33.333333 3.0 0.0 Buy 1 1 0 4 33.333333 5.0 0.0 Sell — Reply to this email directly, [view it on GitHub](#317 (comment)), or [unsubscribe](https://github.com/notifications/unsubscribe-auth/AQ56D3DJKF3ASLE6JOO7HM3UUDD3LANCNFSM5LEDJFYA). Triage notifications on the go with GitHub Mobile for [iOS](https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675) or [Android](https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub). You are receiving this because you authored the thread.Message ID: ***@***.***>

[polakowo]

@Forex2021 direction is controlled by direction argument. Please take time to read the documentation, there are usually small snippets of code under each function to illustrate its usage.