Replies: 1 comment 1 reply
-
I want to backtest last 500 timeframes entries but on a whole df |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi!
Please help me with this problem.
I want to backtest whole dataframe but with entries per 500 timeframes
e.g. len(df)=1500
entry1=df[-500:]
entry2=df[500:1000]
``
coin = 'BNB'
symbol = coin + 'USDT'
cols = ['Open', 'High', 'Low', 'Close', 'Volume']
interval = '1h'
df = vbt.BinanceData.download(symbol, start='30 days ago UTC', interval=interval).get(cols)
df.index = df.index.tz_localize(None)
df.index = pd.to_datetime(df.index, unit='ms')
y = df['Close'].tolist()
y = y[-500:] # last 500 values from a list
x = range(len(y))
best_fit_line2 = np.poly1d(np.polyfit(x, y + np.std(y), 1))(x)
best_fit_line3 = np.poly1d(np.polyfit(x, y - np.std(y), 1))(x)
entries = y < best_fit_line3
exits = y > best_fit_line2
portfolio = vbt.Portfolio.from_signals(df.Close[-500:], entries,exits, init_cash=1000)
Beta Was this translation helpful? Give feedback.
All reactions