When I was backtesting this system using that excel file (attached in my first post of this thread), I did these two things to keep a standard for test (strictly from Nifty's point of view):
1. For previous day's HIGH & LOW, I referred to the EOD chart.
These values are also available at NSE's home page & also in the market-watch window of trading terminal.
I simply noted down these values in the excel file
2. 9:55 to 10:15 data is not available in written format anywhere. They have to be derived from charts. So, for enable back-testing, I had to rely on high & low values in the period till 10:15 from the charts.
3. There's no doubt about the 1st time period at 10:15am. I am not sure, but I think I had misunderstood the theory part about the second time frame period. For back-testing purpose, I fixed the second time period at 10:35 (read somewhere in Dhiraj's reply).
I think that once orders are modified till 10:35am, they will remain fixed & applicable throughout the day. Even if we get new high or low after 10:35, we will not re-modify the order.
I don't know whether I am right or wrong, but I fixed 10:35 as second time interval (to note high & low) and modify orders, as per such new high or low, if there is need.
So, keeping these 3 things in mind, I had done back-testing of this method & I got very good results.
In real-time trading, I will stick to these 3 parameters & don't want to change 10:35 to any other time.
Dhiraj, I may be wrong in this understanding, but now I have become superstitious & would not like to change this mindset.
In short:
a) I refer to EOD chart for previous day's HIGH & LOW.
b) I refer to intraday chart (5min timeframe) only for high & low until 10:15 & then until 10:35.
c) Once orders are modifed / fixed at 10:35, I don't change them at all, even if new high or low is formed.
This is my understanding of the system, and I may be wrong. Dhiraj is the supreme authority, as it is his system.