Part of the investment logic failed to retreat! Sheng Guanda Huang Can： The core of quantifying each strategy is rapid iteration, and the method is to keep testing.

In the context of increasing global inflation, how should investors lay out? How to treat investment opportunities in commodity and stock markets? What is the essence of quantitative money making? How should quantification be iterated quickly?

Recently, China reporter interviewed Huang Can, the founder of Shengguanda. In 2010, the listing of Shanghai and Shenzhen 300 stock index futures triggered a group of practitioners to develop trading systems and trading strategies and set up quantitative investment companies, which can be said to be the first year of domestic quantitative investment. Huang Can is also one of those people. He started his investment career after 2010, mainly engaged in quantitative arbitrage strategies related to stock index futures.

Huang Can regrets that at that time, because of the arbitrage strategy of stock index futures, he basically made money every day, so he didn’t care much about the investment logic of other assets. In 2013, Huang Can established Shengguanda Assets, with a management scale of nearly 7 billion yuan.

After the arbitrage strategy of stock index futures continued to make money for several years, it was finally restricted in 2015. At that time, Huang Can began to turn its attention to assets other than stock index futures, such as stocks and commodity futures, hoping to seek relatively clear and stable investment logic in these two categories of assets.

In Huang Can’s view, quantitative investment is an investment method that applies scientific and technological means to investment. In particular, by using the powerful data processing ability of computers, we don’t pursue every transaction to be profitable, but accumulate investment income through a number of investments with high profit probability. The logic of making money by each strategy of quantitative investment is relatively clear.

For example, arbitrage essentially earns money from short-term market sentiment. When the market is keen on making short-term trend money, the trading opportunities of this strategy become more and the strategic capacity becomes larger. When the market is no longer keen on doing short-term, it is keen on taking long-term, which reduces the fluctuation and makes it more difficult to make money by arbitrage.

CTA strategy is to make profits by paying attention to the price trend of commodity futures in the future. For example, according to the transaction volume, transaction price, the current inventory situation of bulk commodities and the supply and demand situation of upstream and downstream, judge the future rise and fall. Futures can be long or short, so CTA strategy is both long and short. No matter whether the market is in a rising or falling market, CTA strategy has the opportunity to make money as long as the market volatility is at a suitable level.

Among the stock strategies, the largest one at present is index-enhanced products. Because the underlying assets buy stocks, which is easy to be understood by the public, this type of product has been popular since last year.

Huang Can said that the index enhancement strategy mainly earns two parts of money: one part is the money that the index itself rises, and this part is called beta income. Of course, it also bears the fluctuation of decline, so it needs to be held for a long time; The other part is Alpha’s money. At present, different managers have different logic and sources of making money.

Among them, it is mainly divided into fundamental factors and volume and price factors. At present, the mainstream in the domestic market is volume and price factors, because the volume and price factors are much larger than the fundamental factors, and stocks are traded every day, accumulating a lot of data, while the fundamental data of listed companies will not be released every day. Different from the market strategy of quantity and price, Sheng Guanda’s multi-factor stock selection is mainly based on the fundamental quantitative model. Researchers will do in-depth research and analysis of fundamental indicators, and make them into factors by quantitative means, and then combine them with machine learning algorithms to make some combinations. Quantitative investment strategy generally uses historical data to verify the logic behind it, which is essentially that the law of history will last.

It is worth noting that since the beginning of this year, the quantitative strategy has generally retreated significantly, and many products with quantitative index enhancement have even retreated more than the index itself, and the reasons behind it have attracted much attention.

Huang Can said that the main reason is that the investment logic of stock index enhancement strategy products earning excess returns failed this year, including macro-environmental changes and liquidity changes.

"Quantization is not a god, and there is no guarantee that you will make money every year. Therefore, when you lose money, you need to do analysis, modify and iterate your own model. Because of the short transaction cycle, the strategy of quantitative class needs iteration most. " Huang Can said.

Different quantitative institutions have different understandings of how to iterate the strategy. Huang Can said that they need to introduce different perspectives, including macro and medium-sized investment managers, to examine the micro investment logic. Quantification is mostly micro-investment logic, so it is necessary to check whether micro-investment logic is still sustainable. It is not enough from the micro level, and more and higher-dimensional perspectives are needed to help us better understand the market.

In addition to improving the investment logic, it can also limit the scale. Because of the higher frequency strategy, the lower the ceiling of capacity limit.

In addition, Huang Can believes that strategy iteration needs talents with mathematical and physical thinking and strong data processing skills, so as to achieve rapid iteration. At present, the only most effective method in the domestic market is to constantly test, which is equivalent to mining. If the mine is almost dug to the end, it is necessary to constantly explore and dig new things.

Up to now, Shengguanda has developed the main strategies such as quantitative arbitrage, quantitative neutrality, quantitative exponential growth and quantitative CTA. CTA strategy, in particular, has become a rare strategy to make money when the stock has retreated sharply this year.

Under the repeated influence of the situation in Russia and Ukraine and the epidemic, the oil price broke through the $100 mark this year, the price of natural gas soared, the prices of some agricultural products soared, and global inflation continued to intensify.

In Huang Can’s view, the consensus of commodities last year was that it was in a stagflation cycle, that is, the price was affected by the supply side but the demand was not strong enough. Commodities pay attention to the current supply and demand situation from the fundamentals. When the market can objectively reflect the current fundamental situation of commodities, the performance of our fundamental quantitative strategy will be relatively good.

And stock trading is more of an expectation. Huang Can believes that at present, the stock market has reflected the relatively pessimistic expectation of the whole market, including the expectation of the epidemic, but at the same time, we also see that the market as a whole is in an underestimated position due to the various support and steady growth policies introduced by the state at this time, but the investment opportunity should be to do more stocks.

With the gradual layout and investment in various assets such as stocks and commodity futures in recent years, Huang Can himself has become more and more aware of the importance of multi-assets and multi-strategies in the asset management industry. Sheng Guanda, which he founded, is also committed to the steady development of domestic excellent quantitative platform companies.

Editor: Gui Yanmin