From a simple backtest, we can tell that most of the single indicators in the programmatic trading software have failed. 2013 was a very obvious tipping point, with many indicators failing after 2013 or indicators that were not profitable starting to lose more.
Trading strategy systems
In recent years, the number of participants in derivative financial products has increased, with large investment institutions turning to private capital for products such as crude oil and gold. With the widespread use of systematization, globalization, and flattening of the financial system, the profit margin of indicators has been greatly reduced. If the similar logic of a set of systems is used extensively by market traders, the buying and selling orders will squeeze the profitability of the system.
Here we introduce a concept of Strategy over Strategy. The latter is what we generally call indicators. It is mainly used to judge the situation of many spaces. The former is a tool for managing indicators, and we generally refer to the common appearance or law as filtering, positioning, sizing, curve management and risk balance.
Most programmatic traders put their minds to getting a perfect A on the technical analysis level, but perform their intuition on all aspects of B. Subjective intuition brings a lot of parameters, so even if A is perfect, it is useless.
How to judge whether an indicator is failing is also rarely quantified by traders. We should analyze the strategy from a quantitative perspective to see if it has the ability to make a profit, not merely by observation. Here are some ways to rank different strategies in a multi-strategy combination for the way they allocate funds, or to consistently judge whether a strategy is failing.
The overall risk and effectiveness of a comparative strategy can be measured by using the criterion of the benefit-benefit curve. Over a large sample cycle, comparatively better strategies can have similar characteristics, such as profit factors or win rates, at different intervals above the midpoint.
Therefore, the yield curve should present a relatively small upward approach to the right, after processing changes such as commodity price and risk. We can use the yield curve criterion to judge whether a strategy is sustainable or not.
If the average value of the yield curve is subtracted by one standard deviation, then the normal theory is that only about 16% of the daily yield should be below this line. If there is a significant increase in the proportion of option yields below this line, it represents that the earning capacity of the indicator has been compressed or has failed.
In a multi-strategy, we can first optimize the recent indicator and then divide the same parameter value by the long-term indicator. If this value is relatively close to 1, then the strategy is continuously valid.
In multi-strategy combinations, we can use such values to determine how different strategies operate in the commodity, or to allocate funds proportionally to each strategy, adding weight to the currently used strategy.
For example, the numerical calculation of the three strategies is 1.1, 1.2, 1.3, respectively. The former is rated better, and the simple principle of allocating funds can be configured as follows: 0.1 + 0.2 + 0.3 / 0.1, 0.1 + 0.2 + 0.3 / 0.2, 0.1 + 0.2 + 0.3 / 0.3 = 6 : 3 2.
Normally, the odds of winning a leveraged trade are very high. A 50% win rate for a directional band strategy is called a high win rate. For a leveraged model, a win rate of more than 60% is likely to be achieved.
Only the use of capital is less efficient, and therefore the risk is lower. But if the failure of the bilateral or multilateral commodity arbitrage model is the case, then it is obvious and gradual, and not so easily instantly ineffective.
We can judge whether the arbitrage model is failing by simply observing changes in the slope of the yield curve, and once we have a lower level or angle of presentation over the medium term, we can increase it to become a multilateral commodity arbitrage system or convert a commodity portfolio. Each successful addition of a commodity within the portfolio expands the effective space.
Mechanized trading is essentially requiring that every link is running continuously according to the rules we have devised in order to achieve what we expect. Quantitative trading is a derivative of mechanized trading, whereas programmatic trading is based on quantitative trading, giving computers the ability to execute automatically.
So, is it better to digitize each link based on technical analysis? This allows for a more accurate execution strategy first, and then executing a programmatic transaction, which also facilitates further evaluation and feedback, rather than just stopping at the first step in the strategy.
Programmatic traders Programmatic trading and quantitative investing