Hit ratio of the regular EMA crossover (1) and (2) is 32 % whereas it has been observed systematically over 90 % for the periodic compounding and period higher than 8 hours, and even reach 100% for some specific periods. The return is nearly 10 times higher for the periodic compounding than for the other strategies.ДЄfter 50 hours compounding period, the return dramatically drops and then tends to reach the same value as strategies (1) and (2) since the compounding period is probably near the average duration of a period of upward trend. The return even reached a maximum value of 62 at 28 compounding period but the spiky aspect of the curve does not allow to generalize this behavior.ДЄs a comparison, on the same duration, strategy (1) shows a 3,78 return whereas strategy (2) shows 4.88. We expect a maximum stable return of 40 around 30 hours compounding period. Once we reach 9 hours compounding period, both strategy (2) and (1) are outperformed. We observe the periodic compounding perform poorly in a first time, even with a 0,2% loss for the 1 hour periodic compounding. On this graph, we have plotted the results of the total return at the end of the backtesting time lapse versus the compounding period in hours (3), blue curve, with the (1) gray and (2) orange curves. The fees are also included at 0,1% of the trading amount. We use the 01-hs (one hour) chart of the BTC/USDT pair on 657 days randomly chosen, as per the starting time, to avoid potential testerвЂ™s bias. The EMAs are calculated respectively at 13 periods and 48 periods for short and long term (it has actually been proven to be the best EMA combination for crossover strategies). Buy and sell with the total balance periodically between upward and downward crossovers (3).ДЇor the third testing condition, we will perform the backtest from 1 period and up to a number of periods where something significant is observed.Buy with the total balance (capital + profits) at each upward crossover and sell at the next downward crossover (2).Buy with the same amount of capital at each upward crossover and sell at the downward crossover (1).It is well known that such strategy performs poorly and, even on a long-term return, profits are low.ДЇor this study we are going to backtest the EMA crossover for upward trends with: The EMA crossover is what can be considered as the closest of the most basic trading strategy. To have a grasp on what can be achieved with compounding with a real strategy, we propose in the next section to backtest the EMA crossover strategy. If the result is really good here, this is utopian since, in real life, a good strategy will be a compromise between winning and losing trades with risks mitigation and good money management and one can easily guess the devastating effect of compounding on a loosing strategy. We immediately see the effect of compounding the return: the growth switch from a linear slope (no compounding) to an exponential trend and after one year, monthly and daily compounding show respectively higher return by a factor of 1.61 and 1.74. The effect of compounding is shown by the formula: So the profits made are mechanically bigger since the invested capital per trade grows. When trading, one can choose to simply invest the same fixed capital at each trade or reinvest the benefits (or a part) at each trade: this is called compounding. An EMA crossover to an upward trend could be a false signal and leads quickly to the opposite crossover, leading to a downward trend (and ruin the trade). Markets are unpredictable, and the best we can do is calculating a probability band to forecast the price. It looks really simple, but it is not as it seems. If the short period EMA crosses the long period EMA from above to below, the trend is going downwards, and is considered as a sell signal.

If the short period EMA crosses the long period EMA from below to above, the trend is going upwards, and is considered as a buy signal.

Buying or Selling?ДЄt the end it is the only thing that matters, right? A basic observation of the chart gives an immediate answer: This is the basis of the EMA crossover: we look for the point where a short EMA will cross the long EMA. This is a sign the force driving the price evolution are going to reverse the trend: the system pivots due to the momentum created by the divergence of the trends on different time frames. If we have a close look at the graph, we observe sometimes short and long periods EMAs cross.

BTC/USDT at 01-hs chart with Exponential Moving Averages using Superalgos and Sparta data mineДЄs expected, the shorter periods (blue curve at 13 periods as an example) have a better grip on the price whereas longer periods are systematically above or below the price.

20 ema crossing 200 ema