Riding the wind of rebirth

When Mai Xiaomiao raised this question, he did not think about the seriousness of the matter. It was just a trial run process before production. After Zhou Zhi told Li Laosan the story of the future through virtual deduction, Li Laosan began to think about it seriously.

"In fact, the matter is relatively easy to sort out." Zhou Zhi said: "First of all, we must have enough strength to carry out hedging transactions, so that the impact on the market will be minimized;"

“Secondly, we need to pay attention to the scale of the model algorithm. The curve that creates the impact trend should be smooth and should not bring excessive fluctuations to the market;”

“Third, the rules of the local market should be flexible and tolerant enough, and the regulatory measures should be relatively relaxed at the legal level;”

"Then the market volatility itself will be relatively large, so our model will have more chances to trigger a hit;"

"Finally, of course, the flexible inflow and outflow of funds and the various methods make it easier for us to hide our tracks."

"Taking all these into consideration, this is our first choice for the experimental field. By the way, it is best not to expand the scope of the experiment too much. After the system is fully mature, we can gradually connect it to various trading platforms."

"How about the mainland securities market? It's relatively closed and there's no need to alert the enemy. You have a lot of Laojiao and Yilite, which are enough for hedging. The A-share regulatory measures are also relatively primitive..."

"Get lost!" Zhou Zhi was too lazy to say another word.

"Oh..." Li Laosan followed suit and changed his mind: "As for the rest... the Nikkei Index and the Indonesian foreign exchange market are not impossible, but we need to start building positions now and build the pool first..."

"It's definitely too late to do it now." Zhou Zhi thought for a while: "Don't we have a ready-made pool of Hong Kong stocks..."

"Get lost!" Now it was Li Laosan's turn to speak succinctly: "Those pools were left over from previous sweeps... ah no, from bailouts. Hong Kong stocks are very fragile right now. The meeting two days ago also emphasized financial stability. Do you want my dad to kill me for the sake of justice?"

"The dragon-head guillotine is used to behead relatives of the emperor. The tiger-head guillotine you used is almost the same." Zhou Zhi teased, "Why do you still like Peking Opera?"

"Cantonese opera also has Bao Gong's play! My dad is a big opera fan, what's the matter?!"

"Well, I didn't think of that. Forget it, let's get down to business..." Zhou Zhi quickly brought the topic back: "How about... Nasdaq?"

Nasdaq is actually the best target. It was originally the abbreviation of the Automatic Quotation System established by the National Association of Securities Dealers in 1968. Its characteristics are the collection and publication of quotes from securities dealers for over-the-counter non-listed stocks.

Thanks to its flexible mechanism, it often develops new indexes with other partners to timely adjust to emerging forces in the financial product industry that have innovation and profound impact. Therefore, the indexes launched quickly become the focus of world attention.

For example, industrial stock index, banking stock index, computer stock index, telecommunication stock index...

Of course, the Nasdaq 100 Index has the greatest impact on the global financial industry now.

The constituent stocks of this index generally have three characteristics: high-tech, high-growth, and non-financial. This is exactly in line with Zhou Zhi's requirement for AXA Fund to invest in the US financial market. Many companies were preempted by AXA Fund before they even showed the attributes of unicorn enterprises, and participated in them as strategic investors or even angel investors. In the middle, AXA Fund also accurately grasped a large band and now has a large territory.

But this is the ultimate goal, the ultimate vegetable field for this system to reap profits. Before the system is "well trained", if you take it there directly, it will be a disaster if it fails.

Therefore, when Zhou Zhiyi said this, Li Laosan immediately raised these questions.

"Then we will work in two directions." Zhou Zhi said: "The first is to bury ourselves deeper, first considering our own safety and concealment, and ensuring transaction capacity and curve smoothness."

"Then we will continue to keep Baines Financial Technology Company and use its name for all products." Li Laosan made up his mind: "The latter one is also easy to deal with. The main thing is to avoid greed and not think about making all the money."

"Of course it's good that you have this awareness." Zhou Zhi said, "But I'm not talking about that. I'm talking about preventing the risks caused by system bugs."

"Oh, right? I think we can buy the historical real-time trading data of Nasdaq, right? Let's try with those well-known securities market research and analysis institutions. We can bring the data into our own system and create a simulated trading platform to debug our own model."

"We have all these data, but we're just worried that they're not detailed enough." Li Laosan pondered for a moment and said, "I think the five-minute order data and stock price curves in recent years are relatively complete. As for those with a very long history and trend charts within a more detailed time period, it may be more troublesome."

"There is actually a way to do this." Mai Xiaomiao said, "We can use Xiaozhi's knowledge of statistics to simulate more fine-grained data. Although it is not absolutely true, it is at least the most likely truth. By adding mutant genes, we can achieve the purpose of generating training data."

"Can we play like this? How do we do it?" Li Laosan thought it was too magical.

"You don't need to ask too much about this. You won't understand it even if you ask." Zhou Zhi refused to answer: "Anyway, there is a mathematical method to achieve it."

He then added: "If the data is not detailed enough, then we can get more market data to fill the gap. Anyway, the underlying logic of all markets is similar. Connect the data of the London Stock Exchange, Nasdaq, Fed Index, Nikkei Index, Hong Kong Island, and New Island, and we will do T-1 simulation."

"What is T-1 simulation?"

"We build a test system, simulate actual combat, and after the model is trained, we only use yesterday's data as today's real-time data to verify the results. Once it is confirmed to be correct, we will connect it to the real market for operation."

"That's it." Li Laosan stood up excitedly and clenched his fists. "The system transplant and upgrade are up to you. I'll go back today and prepare training data for you. Let's try to get the system out as soon as possible and make money as soon as possible!"

"Why are you panicking?" Zhou Zhi grabbed him and said, "Isn't it just a matter of one phone call for you to prepare the historical data for training? Since you are already here, I can't let you go without showing my hospitality. Besides, after Xiao Miao and I set up the R&D team this afternoon, there will be many details to discuss with you. The communication mechanism with Baines still depends on you."

"Wait until all these preparations are done and the work team can run normally, then it won't be too late for you to leave." (End of this chapter)