Let’s talk about our ETF chart and backtesting

This article will give you an overview of a few ETiFy aspects:

• example of backtesting process
• explanation of why we do backtesting and how we do it
• how we select data for those tests.

Backtesting process

There’s no such token as an ETF. We are using the ticker ETF as a ETiFy solution.

As you see above, a chart shows a % price change during the last 1000 days. And a question may have arisen in your mind — “Why do you test your project on historical data?”.

Before we answer it, you should have two things in mind. First, this is still actual market data so that they can show us some uncommon situations (which is a good factor for testing purposes), plus it’s much easier to prepare them than generated data. And the second one, there are more tests to validate our solution. We plan to do real-time tests on a test environment and spend time with an algorithm to adjust their coefficients.

On the presented chart, there are tokens with final return:

• BTC x2.5
• ETH x7.7
• XRP x1.9
• ADA x7.7
• BUSD
• ETF x4.5

You might think, “he wants to convince me of his project and show me results where his solution is not the best to choose.” And you are partially correct. Yes, it’s not focused on maximizing gains but on lowering the risk.

Let’s go deeper

Our ETF token price is established on the price of underlying tokens. Each token should have a value close to 20% of the portfolio. Our algorithm will execute a rebalance action once daily when this ratio changes. ETF token price will form on circulation supply and TVL, but we will tell you more about it in the following article.

So when we know on a high level how it works, we should talk about assumptions before the result explanation. Most of them are built in our early project stage.

• No transaction fee. We want to start on an some layer 2 so that those fees will be minimal so that we will ignore it for now.
• No transactions during this period, just initial investments. This one is important. We decided to skip the ETF transaction because it should not affect our final percentage result and to avoid stable coin liquidity problems during the test. We will cover this issue more broadly in the other article.
• Selected tokens were projects with solid fundamentals 1000 days ago, and someone may pick them for that kind of ETF then.

Wen moon, sir?

Our results show that our project can generate about x4,5 returns. Why can’t I buy ADA and hodl or, even better, sell on top? Of course, you can, but all investors know it’s more complex. First, you need to find a project and predict its growth, then this project must follow your prediction, and in the last phase, you need to find a top and convince yourself to sell.

It sounds easy to gain x60 like that. But the truth looks a bit different. And we want to avoid fighting with reality. We want to accept it and adjust our strategy to it.

We do backtest on multiple ETF bags to check the effectiveness of our solution. Here we can see one example. A period is only 700 days because of one project lifespan. In this case, our ETF is on a giant loose x0,347 but still is a top 3 result in this token stack.

If an investor decides to pick one token and go complete with it, he has a significant risk of losing everything (we wait to talk about features now). Smart ETF will allow him to invest similarly by buying one token with exposure to multiple tokens. If one of these tokens collapses, the portfolio loses only part of the balance. An investor can forget about his investment for some time because our algorithm will manage his portfolio daily without additional fees.

Why do we do these backtests?

Our main goal is to check what results from our algorithm will give us compared to other tokens. And the second idea behind that is to adjust the coefficient we use in the algorithm.

The next test phase will be on the mumbai testnet to check market behavior. Stay tuned.