Cyatophilum Smart rebalance strategy

2021년 11월 17일

An indicator to backtest and create automated Smart Rebalance strategies.
The strategy tries to beat holders while keeping your portfolio repartition stable.
Configure the indicator to receive alerts for several pairs at a choosen rebalance time frame.

═════════════════════════════════════════════════════════════════════════

█ HOW IT WORKS

Here is an example:

The user chooses a certain number of pairs, up to 4 pairs for now. An initial capital, a fees % and a rebalance time frame.
The capital is split equally between each pair. (ex. for 2 pairs: 50% on each).
On each rebalance time frame (on the bar it happens), pairs balance who grew bigger than 50% will be redistributed to other pairs, making sure that after the transaction, the total capital is once again split equally through all pairs. This means selling pairs whose price went higher than the others, and buying those who dropped. In other words, selling high and buying low.

On each interval, the strategy will send orders for each pair saying whether to buy or to sell, and the quantity.

█ HOW TO USE

Creating a strategy

When adding the indicator to the chart, you must confirm the pairs you want to use, the rebalance time frame, the initial capital and commission fees.

The inital capital corresponds to the quote currency of all pairs, so each pair must have the same quote currency. (ex: USDT or BTC markets)

It will also be asked to select a "Strategy Start Date", which is the starting point of your strategy. Select the current bar if you want to start a strategy and get accurate order quantities. Otherwise, select a bar in the past for backtesting.

When setting the "Strategy Start Date" to the current day, it will look like this:



The strategy equity is equal to the intial capital, and no order has happened yet, the backtest panel shows 0%.

The first rebalance will happen at the next interval: here the 15th november.



Since this is the first rebalance, the strategy buys each pair with the same quantity in quote currency value.
After 1 rebalance, the strategy equity drops a little, since we only bought and paid fees.

Backtesting

Since the strategy is trading several pairs, the Strategy Tester cannot really be used. This is why the indicator has a backtest panel and a equity line plot that shows the progress of the strategy.


This view can be replicated by disabling the labels in the style tab.

You can also use the Data Window to track the strategy pairs balances on each bar.

Creating Alerts

The indicator is using the alert() calls: it only uses 1 alert slot to send order messages for each pair.

To set your alert messages, open the indicator settings.

You can use the placeholders {pair}, {order_direction} and {quantity} to get their values replaced by respectively the pair ticker (symbol name without the exchange prefix), the order direction "BUY" or "SELL", and the quantity amount in quote currency.

Create your alert after you set the messages in the indicator settings, and make sure "Any alert() function call" is set in the alert option.


█ LIMITATIONS

 • Small quantities
An order with a too small quantity might not get executed by the exchange. To increase the quantity of the strategy orders, either increase your strategy capital, increase the rebalance time frame, or reduce the number of pairs.

 • Not enough data
The strategy will give an error if one of the pair data does not go back to the strategy start date. Typically if a crypto pair is new and was added to TradingView recently, you will not be able to backtest prior to that.

 • Chart and rebalance Time Frame
The strategy will give an error if one of the two conditions are not met:
- The Rebalance time frame must be greater than current chart's time frame.
- The Rebalance time frame must be a multiple of current chart's time frame.

Please read the author instructions below for access and automation.

2022년 6월 20일

릴리즈 노트

Updated fees calculations
Added a configurable quantity % allocation for each pair

2022년 8월 22일

릴리즈 노트

Updated alert inputs