Forums  > Pricing & Modelling  > Facing with passion, boring and lonely for more than ten years, changing from one thousand to one million is not a dream!  
Page 1 of 1
Display using:  


Total Posts: 1
Joined: Feb 2022
Posted: 2022-02-03 10:29

Focusing on this international common currency transaction with hedging function, the backtest of programmatic trading is as follows:

1. The initial 1000 dollars will change to 48000 dollars in five years from 200901 to 201401. Not bad!

2. Start 45,000 dollars from 201401~201901 for five years, I rub it, I hardly make any money, and it still stays at 48,000 dollars. It’s really hard to make money! It's so hard!

3. Start 45,000 dollars from 201901~202106 for two and a half years. Unexpected!

Summary: This automatic trading program belongs to a complete trend trading, and the holding time is long or short. According to the trend, the trading rules of trial and error are strictly implemented. Strictly stop losses, and don’t be discouraged by losses; strictly shift losses, make profits without fluke, and earn without greed.

From more than ten years of backtesting, it is found that the quantification of K-line, moving average and trend channel is the most fundamental, that is to say, K-line must be 100% of the real process of volume and price changes. In layman's terms, the daily line is the daily line, H1 is the H1, and H4 is the H4. Some platforms have 6 daily K-lines a week, which completely distorts the working calendar and the shape of the K-line. Some platforms have only three H4s on the day. Hours of transaction time, bullshit! And so on, when you quantify into program trading, no matter the K line or the corresponding moving average, the quantification must be a serious mistake. Even if the trend channel is generally consistent and does not affect the position closing or stop loss and profit, your order point must be seriously dislocated due to quantitative errors, resulting in your final trading results being very different.

Therefore, according to the opening and closing time of gold, the division of servers and K-line time of such platforms as IC, XM, Inno, etc., may be more appropriate as the basis for trend quantification. Therefore, the K-line charts of many platforms are "out of context" and cannot be used as a quantitative basis at all! Remember! Otherwise, wrong quantification leads to wrong trend conclusions and results that you did not discuss!

For details, see:
Previous Thread :: Next Thread 
Page 1 of 1