Categories
Crypto-ML News

Intelligent Crypto Investing with Machine Learning

Last Updated on November 23, 2021 by admin

At Crypto-ML, we’re pleased to announce our new machine learning-based crypto investing strategy: CML-I.

This strategy uses machine learning to determine longer-term entry and exit points in order to capture bull and bear market cycles. Our other strategies fall into the swing trading category, making CML-I the first strategy to approach the market from an investor’s perspective rather than a trader’s.

Why Investing Matters

As a bit of a personal story, I can share that I initially became interested in crypto because of its intense volatility. If volatility has some degree of predictability, it can be exploited as a way to make money as a trader. The underlying asset doesn’t matter as long as it has this “reasonably predictable volatility” feature.

In order to understand the drivers of volatility and gain an edge as a trader, I educated myself on the crypto space.

  • What causes price movement?
  • How can cryptocurrencies be valued at a fundamental level?
  • What causes one to succeed over another?
  • Why are they so rapidly appreciating in terms of market capitalization?
  • Where is the money coming from?

Here’s our take on why Bitcoin is valued so highly.

A funny thing happened along the way. Rather than seeing crypto as an asset to exploit, I began to understand its revolutionary potential.

Decentralization, in particular, has the ability to completely upend many of the foundational business and organization models in existence today. If you missed out on the early stages of Web 1.0 and Web 2.0, now is your chance to be part of Web 3.0.

In fact, most of the Crypto-ML team went through a similar personal journey. The thought of intelligently investing in crypto and decentralization has become an obsession.

As a first step toward Crypto-ML becoming a platform to provide actionable AI-driven crypto investing insights, we are extremely proud to release CML-I: Investing Machine Learning.

Investing Compared to Holding and Trading

Based on what I said above, you may wonder how crypto investing, holding, and trading differ

  • Trading is shorter-term in nature. On Crypto-ML, this means swing trading, where trades may be open for hours to days. In this category, you usually don’t care about the underlying asset. You just want the volatility and some understanding of its drivers.
  • Holding an asset means you keep it indefinitely without a planned exit strategy. Oftentimes, it’s best to make these funds hard to withdraw. You likely have a very long time horizon for this part of your portfolio.
  • Investing falls somewhere in the middle. You believe in the fundamentals of an underlying asset, but realize nearly everything moves in cycles. That means even though you expect the asset to grow over the long run, you also expect there to be bull and bear markets. By actively navigating these boom and bust cycles, you expect you can outperform “just holding” the asset.

Nearly everything moves in cycles.

In fact, in the Market Index video below, I walk through how following very basic crypto investing rules, you can greatly outperform holding. The rules in this demo are simple:

  1. Exit during mania as defined by +80 on the Market Index.
  2. Enter during panic as defined by -120 on the Market Index.

That’s a very unsophisticated approach, but it still outperforms hold by about 300%.

Hold, Invest, and Trade with Crypto-ML’s Portfolio Management

Since Crypto-ML’s Auto Trade feature is actually a Portfolio Management tool, you can assign percentages of your crypto portfolio to different strategies, including hold, investing, and trading.

Based on your risk appetite and your general outlook, you can craft the right portfolio for you.

Crypto Investing with Crypto-ML Portfolio Management

Other Tools for Crypto Investing

While CML-I is our first trade strategy for investors, all Crypto-ML members have long had access to the Market Index which has been designed from the outset to help investors understand the bigger-picture view of the market. It seeks to quantify features of bull and bear markets, such as:

  • Overbought vs oversold
  • Exuberance vs despair
  • Greed vs fear
  • Speculation vs investing

The good news is CML-I consumes all of the same data the Market Index does.

However, since machine learning drives CML-I, there are no hard and fast rules programmed in. For example, CML-I does not “sell when the Market Index is over 80.” Rather, it considers the same ingredients and comes to its own conclusion as to when to issue an alert. Having a rule such as “when it’s over 80” is too simplistic. We want machines to sift through vast amounts of data and identify patterns we likely are unable to find on our own.

How CML-I Works

CML-I is based on the same technology foundation as CML-A. That means it utilizes the same prediction, anomaly detection, and trade mechanics. And as noted above, CML-I also gathers the same data as our Market Index.

It’s a thoroughbred machine learning system.

Put together, it’s considering the following types of cryptocurrency market inputs:

  • Exchange data
  • Technical indicator data
  • Social sentiment and volume data
  • Search sentiment and volume data
  • Bitcoin dominance data

Learn more about this approach in our Machine Learning Technology Guide.

What You Can Expect

If you’re familiar with CML-A, you can expect similar behavior from CML-I, just with lower sensitivity. It utilizes a longer-term perspective in its decision-making, helping to avoid the choppiness of intra-day cryptocurrency markets. CML-I is designed to seek out and capture longer bull runs.

In practice:

  • CML-I generates about a quarter the number of trades that CML-A does.
  • Like with CML-A, CML-I will exit markets when it senses trouble and will stay out during larger downtrends.
  • The drawdowns CML-I has displayed thus far have not been no different than CML-A.

Based on the time CML-I has been running behind the scenes, it has significantly outperformed our other strategies.

That said, there are some considerations:

  • Profit is not booked as frequently as with CML-A. This may result in drawdowns and oscillations in positions.
  • Fewer trades means less feedback, so it may take longer to know if CML-I is working the way you want.
  • During sideways or down markets, CML-I may stay silent and keep you out of the market completely. While this may be a bonus as it reduces risk, it also means CML-I may miss some smaller opportunities.

As a general best practice, it’s good to select more than one strategy when using Auto Trade.

Conclusion

CML-I is an initial step in our journey to bring more crypto investing-focused tools to Crypto-ML. If you’re ready to get started with CML-I:

About Crypto-ML

Crypto-ML provides machine learning for crypto traders and investors. Gain crystal-clear signals and deep market insights. Add predictive capabilities to your toolbox. Learn more and join for free.

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Add predictive capabilities to your crypto trading and investing.Join Crypto-ML for Free