Shulou(Shulou.com)05/31 Report--

This article mainly introduces python how to achieve linear hanging single-flow strategy related knowledge, the content is detailed and easy to understand, the operation is simple and fast, has a certain reference value, I believe that after reading this python how to achieve linear hanging single-flow strategy article will have a harvest, let's take a look.

How to use data playback

The data playback function can be trained without exchange trading time restrictions, support a variety of commodity futures and digital currencies, the market can be played back manually or automatically, and the historical market start and stop time and playback speed can be set freely. Compared with other software generally use K-line data playback mode, inventors quantify the use of Tick-level data playback mode, really close to the real trading test environment, reproduce the opening price data, so that traders immerse themselves.

Open the inventor's quantified official website (fmz.com) to register and log in, and then click the data exploration in the control center to present the data playback function page. It * * has four option boxes and a select button. First, click the select button to display only the varieties that support real disk playback, then select the variety to be played back in the upper left, then select the start and end time of the data in the latter two option boxes, then select the time period of the data for real disk playback, and finally click the rightmost Go button to turn on the data playback function.

It is divided into three pieces under the data label. On the left is the transaction history, where all the closed orders are shown in chronological order. In the middle is the sale of 20 files of depth of inventory data. On the right is the control area for data playback, where you can choose manual and automatic data playback, which is as simple as using a media player.

The position index can drag the cursor back and forth to quickly select the start time of data playback.

At the bottom, you can also control the playback speed of the data by moving the cursor to the right and left, which can both accelerate and decelerate when the data is played back in milliseconds.

Build strategy logic

Although there are many factors that affect the rise and fall of prices, including the global economic environment, national macro policies, relevant industrial policies, the relationship between supply and demand, international events, interest rates and exchange rates, inflation and tightening, market psychology, unknown factors, and so on, but the final price on the market is the result of competition between many parties and short parties. If there are more buyers than sellers, the price will rise; on the contrary, if there are more sellers than buyers, the price will fall. Then we only need to analyze the price to make a trading strategy.

Through the inventor's quantitative replay of the btc_usdt trading pair of the Yuan'an Exchange in the last month, we find that when the market rises and falls sharply, the order volume of the Tick data is obviously asymmetrical. When the market rises sharply, the hanging order quantity of the bulls is obviously larger than that of the bears; when the market falls sharply, the hanging orders of the bears are obviously smaller than that of the bears. So can we predict the price rise and fall in a short period of time according to the number of orders in the order book?

The answer is yes.

By collecting deep Tick data, we can calculate and compare the pending orders between long and short parties. If there is a great disparity between long and short combined orders, it may be a potential trading opportunity. For example, when the amount of long orders is N times that of short orders, we can think that most people in the market are bullish, and the probability of price increases in a short time in the future will increase; when the number of short orders is N times that of bears, we can think that most people in the market are bearish, and the probability of prices falling in a short time in the future will increase.

Write a trading strategy

According to the above strategy logic, start to implement it in code. Open: fmz.com website > Log in > Control Center > Policy Library > New Policy > Click the drop-down menu in the upper right corner and select Python language to start writing policies. This strategy serves as a valuable tool for teaching, so I try to simplify the strategy and pay attention to the comments in the following code.

Step 1: write a strategy framework

# Policy main function def onTick (): pass# program entry def main (): while True: # enter infinite loop mode onTick () # execute policy main function Sleep (1000) # hibernate for 1 second

When we write a strategy, we should write from uppercase to lowercase, just like building a house, building frames first and then walls. In this framework, we use two functions: the main function and the onTick function. Where the main function is the entry of the program, that is, the program will be executed from here, and then enter the infinite loop mode, repeatedly execute the onTick function. Then we just need to write the policy content into the onTick function.

Step 2: write global variables

Vol_ratio_arr = [] # multiple pending orders ratio array mp = 0 # virtual position

The reason why vol_ratio_arr is defined as a global variable, because my strategy needs to collect a piece of Tick data multi-empty pending single ratio, if we put the vol_ratio_arr variable into the onTick function, with the loop running is obviously unreasonable, what we need is in the loop mode, when a certain condition is reached to change the value of the variable, the most reasonable way is to put this variable outside the loop.

Position management is very necessary because it is related to the buying and selling logic. in general, we calculate the currency pairs held by obtaining accounts in spot transactions. In order to simplify the code, a global virtual position variable is directly defined to control the buying and selling logic.

Step 3: calculate the current long-to-short ratio

Depth = exchange.GetDepth () # get depth data asks = depth ['Asks'] # get selling price array bids = depth [' Bids'] # get buy price array asks_vol = 0 # all selling orders bids_vol = 0 # all buying prices for index Ask in enumerate (asks): # ergodic selling price array # Linear calculation of all selling order asks_vol = asks_vol + ask ['Amount'] * (20-index) for index, bid in enumerate (bids): # ergodic buying price array # Linear calculation of all purchase price hanging order bids_vol = bids_vol + bid [' Amount'] * (20-index) bidask_ratio = bids_vol / asks_vol # calculation of long-short ratio

As we all know, digital currency is usually 20 files of in-depth data, so we can add up the hanging orders of bulls and bears to calculate the ratio of bulls to bears. When this value is greater than 1, it indicates that the price will rise in a short time in the future. When this value is less than 1, it means that the bears are larger than the bears, indicating that prices will fall in a short time in the future.

But there is one thing that needs to be distinguished. The closer the hanging order is to the opening, the stronger the willingness to be bullish or bearish. For example, the purchase order in the first gear must be stronger than the buy order in the 20th gear. Therefore, when we accumulate hanging orders, we need to give different weights to the 20-file hanging orders in a linear way, which will be more reasonable.

Step 4: linearly calculate the long-to-empty ratio over a period of time

Global vol_ratio_arr, mp # introduce the global variable vol_ratio_arr.insert (0, bidask_ratio) # put the multi-empty ratio in the global variable array if len (vol_ratio_arr) > 20: # if the array exceeds the specified length vol_ratio_arr.pop () # delete the oldest element all_ratio = 0 # temporary variable, all multi-empty pending order ratio all_num = 0 # temporary variable All linear multipliers for index, vol_ratio in enumerate (vol_ratio_arr): # variable global variable array num = 20-index # Linear multiplier all_num = all_num + num # Linear multiplier cumulative all_ratio = all_ratio + vol_ratio * num # all multi-empty hanging order ratios accumulate ratio = all_ratio / all_num # linear multi-empty hanging single ratio

Dividing the long accumulation order by the short accumulation order can get the long-short ratio, but this is only a Tick data, if only one Tick data is used to determine buying and selling transactions may not be a wise choice, because in a rapidly changing market, a Tick data is not convincing. So we need to collect a fixed piece of Tick data, and then calculate a fair value linearly.

Step 5: issue an order transaction

Last_ask_price = asks [0] ['Price'] # the latest selling price, the latest buying price last_bid_price = bids [0] [' Price'] # the latest buying price, the selling price if mp = = 0 and ratio > buy_threshold: # if no money is currently held And the ratio is greater than the specified value exchange.Buy (last_ask_price, 0.01) # Buy mp = 1 # set virtual position if mp = = 1 and ratio < sell_threshold: # if the current currency is held and the ratio is less than the specified value exchange.Sell (last_bid_price, 0.01) # sell mp = 0 # reset virtual position value

Because we need to specify a price when placing an order, we can directly use the latest selling price when we buy and use the latest buying price when we sell. Finally, at the end of the order transaction, reset the value of the virtual position.

This is the end of the article on "how to achieve linear single-stream strategy in python". Thank you for reading! I believe you all have a certain understanding of "how to achieve linear single-stream strategy in python". If you want to learn more, you are welcome to follow the industry information channel.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.