Discussion of article "Comparison of different types of moving averages in trading"
Описание торговой стратегии
A simple strategy with obvious market entry and exit conditions was chosen for testing the indicator.
Market entry conditions.
- Preliminary signal to buy: the indicator line crosses the body of a "bullish" candle. Further, if the difference between the current and previous values of the indicator is greater than the specified parameter Growth factor (indicator is growing), we open a deal to buy.
- Preliminary signal to sell: the indicator line crosses the body of a "bearish" candle. Further, if the difference between the previous and current values of the indicator is greater than the specified Growth factor parameter (the indicator is falling), open a deal to sell.
Market exit conditions:
- on achievement of TakeProfit or StopLoss levels;
- if a Buy trade is opened and the indicator line crossed the body of a bearish candle;
- if a Sell trade is opened and the indicator line crosses the body of a "bullish" candlestick.
Questionable TS for the conclusions drawn in the article.
The article says "Testing was performed for the period from 01.01.2016 to 09.09.2017."
And for what period was the optimisation performed? The results are too good for muvingas.
Optimisation was performed for the period from 01.01.2016 to 09.09.2017. I did not overestimate the results, there is no point, it's not like I'm selling a trading robot on muwings.
Great )))) First on this period we do parameter fitting, and then test on the same period )) Try to do forward optimisation and testing, you will be unpleasantly surprised.
Moving averages are probably a good way to analyse natural transients (e.g. seasonal temperature changes).
How can some bar, whose value is the average of the last few bars, show where the price will go?
The accuracy is of the sort - it went up, so it will continue to go up.
Great )))) First we do parameter fitting on this period, and then we test on it )) Try doing forward optimisation and testing, you will be unpleasantly surprised.
Yes, you are right. Back optimisation and forward tests would be more realistic. I did such optimisation and testing for EURUSD for TEMA, NRMA and DEMA (they gave the best results in the article).
Back-optimisation. Period 01.01.2016-04.11.2016 (half of the time interval on which I tested earlier).
Forward testing. Period 05.11.2016-09.09.2017.
The results are summarised in the table:
| Indicator name | Values of optimised parameters | Net profit (optimisation) | Net profit (testing) | Recovery factor (optimisation) | Recovery factor (testing) | Profitability (optimisation) | Profitability (testing) | Sharpe ratio (optimisation) | Sharpe ratio (testing) |
|---|---|---|---|---|---|---|---|---|---|
| TEMA | Period - 44, Growth factor - 0.0002 | 829.62 | 1072.76 | 2.39 | 3.93 | 1.32 | 1.41 | 0.1 | 0.13 |
| NRMA | Period - 10, Growth factor - 0.0001 | 772.78 | 415.26 | 1.49 | 0.89 | 1.26 | 1.15 | 0.09 | 0.05 |
| DEMA | Period - 49, Growth factor - 0.0002 | 541.22 | 575.92 | 1.21 | 1.68 | 1.24 | 1.31 | 0.09 | 0.10 |
For NRMA, the testing turned out worse than in the article. TEMA and DEMA showed good results.
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
New article Comparing different types of moving averages in trading has been published:
This article deals with seven types of moving averages (MA) and a trading strategy to work with them. We also test and compare various MAs at a single trading strategy and evaluate the efficiency of each moving average compared to others.
Let us consider displays of different variants of Moving Average indicator at a price chart. Figure 1 demonstrates variants of Moving Average indicator with the period of 12, as calculated by Close prices.
Fig. 1. Variants of Moving Average indicator
Author: Aleksey Zinovik