On this article, we’ll analyze a particularity of the cryptocurrency market, particularly the correlation that exists between the value evolution of Bitcoin (BTC) and that of Ethereum (ETH), the principle altcoin in the marketplace. We’ll attempt to exploit this statistic to create a scientific buying and selling technique, engaged on the Ethereum-Bitcoin (ETH-BTC) pair.
The attribute that we want to analyze is linked to an alleged recurring habits (in any other case referred to as Bias) which might present itself through the week: we need to test whether it is true that it’s sensible to carry Bitcoin (BTC) for the start of the week till the weekend, and convert it into Altcoin (subsequently into Ethereum on this instance) through the weekend, when certainly Bitcoin appears to take a break in favor of the Altcoin which is somewhat appreciating.
The notion of bias, that’s to say a recurring habits of costs over time, is among the easiest triggers on which it’s doable to construct a buying and selling technique. It is a market inefficiency that recurs with a sure regularity, for instance at particular occasions or days of the week. When it extends over longer durations, we usually converse of seasonality.
Correlation between Bitcoin and Ethereum: value evaluation over time
To test if this recurring habits between Bitcoin (BTC) and Ethereum (ETH) actually exists, we’ll use in a preliminary section the Bias Finder, a instrument developed internally on the Unger Academy to simplify the seek for bias on futures. The Bias Finder, the truth is, primarily based on historic knowledge of a sure instrument, can shortly analyze traits on candles of various length between 5 and 60 minutes. As soon as the instrument and the interval have been chosen, it’s then doable to guage totally different time horizons: every day (Intraday), weekly (Weekly), month-to-month (Monsuel) or annual (Yearly). For every of them, a chart is generated that reveals the development of the chosen instrument, expressed as a mean financial tour over time in absolute or proportion phrases.
In our case, by choosing the Ethereum-Bitcoin (ETH-BTC) pair, with a time slot of quarter-hour and historic knowledge from 01/01/2018 to 12/31/2024 (knowledge supply: Binance), we are able to shortly assess whether or not the habits that was assumed through the week is current or not, by analyzing the development of the typical financial tour of the instrument.
From the weekly chart proven in Determine 1, the place every line represents a yr (from 2018 to 2024) and the darkish blue line represents the typical for the whole interval, we are able to see that statistically, excluding 2021 (inexperienced plot) , the Ethereum -Bitcoin pair reveals some downward development (the value of Bitcoin will increase relative to that of Ethereum) from the start of the week till round Thursday, then rises barely through the weekend.
The bias, though not very pronounced, subsequently nonetheless appears current, we might subsequently hypothesize a scientific buying and selling technique to guage its profitability.
Logic of the Bias buying and selling technique on the Ethereum-Bitcoin pair (ETH-BTC)
Initially, utilizing a 60-minute Ethereum-Bitcoin chart and referring to the UTC time zone, one might hypothesize promoting (i.e. shopping for Bitcoin) on Monday at midnight (night time between Sunday and Monday) and in addition purchase Ethereum (ETH) on Friday. at midnight. Any open place would subsequently be closed by opening the place in the other way, with none cease loss or take revenue orders.
To simplify the studying of the outcomes, the operations will likely be carried out with 100 Bitcoin (BTC) for every transaction, thus permitting a direct analysis of the outcomes as a proportion of the capital employed, which is exactly 100 Bitcoin (BTC). Figures 2 and three present the metrics for this quite simple technique, with none sort of filter utilized to the operations.
It’s instantly noticeable to what extent the expansion of shares confirms the presence of the bias recognized with the Bias Finder, though the development is somewhat unstable within the historic vary thought of.
Evaluating the typical commerce, we discover a worth of 0.38 Bitcoin (BTC), or 0.38% of the capital employed (100 Bitcoin), which is also an enough worth to cowl operational prices, on condition that It is a tough technique, however it could definitely be improved.
Technique optimization on Ethereum-Bitcoin: methods to enhance efficiency
We might first optimize the timing of entry into the market by various the times and time of entry: for the lengthy entry day between Wednesday and Saturday, and for the brief entry day between Sunday and Tuesday. Determine 4 reveals the results of the optimization carried out with MultiCharts, the place the times of the week vary from 0 (Sunday) to six (Saturday). It’s value noting that lengthy entry (shopping for Ethereum) on Wednesday at 4:00 p.m. is more practical than on Friday, as beforehand assumed. The brief entry can, nonetheless, be left on Monday at midnight, as a result of advancing or delaying it by one hour doesn’t considerably change the consequence.
The subsequent step might be to guage the usage of cease loss and take revenue to raised handle open positions. By optimizing the cease loss between 0 and 5 Bitcoin (equal to 0% – 5% of the open place), in steps of 0.1, and the take revenue between 0 and 10 Bitcoin in steps of 0.25, you get hold of the combos in Determine 5., amongst which you’ll be able to go for the one with Cease 1.9 Bitcoin and take revenue 7.5 Bitcoin, which ends up in a superb internet revenue/max drawdown ratio and a mean transaction that quantities to 0.55%.
As Determine 6 illustrates, the fairness line has considerably improved, as have all of the technique parameters noticed beforehand.
Nonetheless, analyzing lengthy and brief trades individually, we are able to see that the lengthy facet is the least environment friendly. We might subsequently think about filtering lengthy entries utilizing value fashions, to solely function when situations are extra favorable for the event of bias. this one is making an attempt to roll. Nothing would stop you from doing the identical on the brief facet, supplied you don't filter an excessive amount of and threat falling into overfitting the technique.
Conclusions and concepts for creating a buying and selling technique primarily based on Bitcoin-Ethereum correlation
The bias recognized between Ethereum and Bitcoin represents an attention-grabbing instance of market inefficiency that may be exploited with a scientific strategy. Regardless of the simplicity of the preliminary strategy, the outcomes spotlight vital potential, significantly after optimization of commerce entry and exit guidelines.
This research goals to be a place to begin to additional discover the topic and refine the technique to additional enhance its efficiency. The chances for enchancment are quite a few, and it’s as much as the reader's curiosity and initiative to discover them to create more and more efficient programs.
See you subsequent time and blissful buying and selling!
Andrea Unger
