so I was trying to backtest on Apollo Hospital historical data for swing trading, and guess what I found; Kite’s own historical data is a mismatch with NSE historical data itself. Even tradingview chart is also synced to NSE historical data.
My way of backtesting is, whenever I have to enter or exit a trade for swing long, then I pick opening price of next candle. So, here entry price and exit price both are the opening of next candle and that time is also from next candle too.
Above image is a trade result from one of all results which I was trying to backtest on KIte’s historical data.
This image is NSE own charting data from that same candle.
You all are seeing time diffrence of same candle because NSE calculate time in a different way. And that way is pretty correct. NSE charting
Now my question is with rest all of you [whoever in the Algorithmic Trading or Quantitative part] plus zerodha core developer team who is trying to manage the data on their end.
Am I not exaggerating much? or Am I right to ask this question?
This above drawback pulls something critical from the zerodha development team that they lacks data verification process from NSE.
NSE data is core so we have to comply ourself according to them, not on our own.
What is going wrong on your end Zerodha? Why can’t you manage data in proper sync with NSE?
What if? we found these types of mismatch from most of the price chart at zerodha’s end?
Unless you participate in preopen, you dont really get open price anyway.
There is always some difference between backtested prices and what we can actually get. Even with bar close, you cant really execute to always get the closing price. A close price could have been either bid or ask at that time, and further even if aligned with your direction it may not have enough volume and you cant really execute at exact closing time. Slippage is fact of life. Best we can do is to try to estimate it.
I would not worry too much about this, but make sure your edge is more than enough to overcome things like this and expenses too. Best way to estimate is start trading live with small qty. That will give some rough idea + slippage will increase once you increase size, so its an ongoing thing. Stock volume also can change with time.
(Note: I say “might”, as the linked thread doesn’t specifically state anything about 15min candles being reviewed in this thread. Also the linked thread is somewhat old, so not sure if things have changed in the meantime)
