Artificial Intelligence – New City Trader https://newcitytrader.com Magazine for Modern Traders: AI, alternative data, automats, hybrid systems Fri, 22 Dec 2023 17:46:32 +0000 en-US hourly 1 https://newcitytrader.com/wp-content/uploads/2021/09/cropped-Logo-nct-a-32x32.png Artificial Intelligence – New City Trader https://newcitytrader.com 32 32 The Lowest Hanging Fruit [Part 2] https://newcitytrader.com/the-lowest-hanging-fruit-part-2/ Tue, 02 May 2023 07:00:00 +0000 https://newcitytrader.com/?p=8632

We present the promised second part on the use of AI tools in trading. In it, we show how the new tools can be used in an example fund. It consists of discrete traders and the head is a trader who has practically been trading oil for several decades. Part I can be found in the first issue of our magazine

The Fastest Path To Cash: Support The Best And Improve What Already Works

Let’s go back to the fund and ask yourself – what tools to use and where. This is the question of where investments can bring the fastest, measurable benefits: stop the erosion of profits, reduce risk and perhaps increase returns.

What I am describing now is not a universal prescription, because each situation may be different. From my perspective, I provide topics to think about. I would say this – the general rule is that we should support what works best, brings the best profits and what carries the most risk. In the case of a fund, or discrete traders – this is to support the best traders.
Who in our fund is the best?

In the described fund, the best trader is the holder. Support for him is obvious as long as he wishes, of course. However, AI support in this case can fail, as we are about to demonstrate. We get the best practice from implementing AI tools when we support, improve, automate what already works well. In general, we will get the best results by supporting what works best.

If we have a group of top traders then they are a fairly obvious choice.

However, one more condition must be met – they must be open to new methods. It happens, for example, that the statistical approach is so far from the trader’s experience that he is unable to switch. He is unable to combine the insight provided by the new knowledge with what he already knows. The result is unnecessary tension, and investment and time are lost.

The best usually don’t have a problem with this – that’s one of their key characteristics. They are open to change, they know that the market is constantly changing and you have to adapt. As an example, let me mention Larry Williams, who I interviewed for a book a few years back. He is 77 years old today and you can find his current lectures on the Internet about the new indicators he is creating. Age is no barrier to taking advantage of new things. Our trader, the head of the fund is a man in his 70s and yes, I think he is the one who should be supported if he wants to.

The second question is – where is the greatest risk and how can new tools reduce it?
The third area is (and here I think there is very great potential) the implementation of new MM (money management), I will write about this in future articles.

And some more tips at the end

If you are thinking of a larger AI department (using only new tools) then create a separate team. In the next issue I will describe good and bad practices in this area, i.e. how not to lose tens of millions using what others have already burned on.

New Strategies For Managing Position Size For Different Classes Of Signals

I often refer to what the best discrete traders do because it is their practices that are worth emulating and improving (through automation).

The best ones adjust the size of the position according to the “goodness of the signal,” in the sense that they are willing to increase the position when the signal is exceptionally good (according to their evaluation criteria).

If we start from the point that some signals are better and some are worse then we can move on to the creation of a seamless MM, based, for example, on statistical evaluation, assessment based on the results of the application of Machine Learning.
The starting point can be a simple observation that the best inputs may have something in common. What is it, what are the factors – this is the task for the ML tool (classification of inputs due to various parameters).

If we find this, then we can count new dynamic rules for allocating position size to each signal ( depending on the parameters that characterize the input – whether it is “better” or “worse – to which class it belongs).

This approach opens the door for us to create very liquid MM strategies – the best signals get the most and are pyramided and the weakest (but still “valuable”) – get less or only “control inputs”.

In the last ones, for example, we can try “control entries” and if the position holds and is secured – we add so much that if the market goes against us – we will come out at zero (this is one of the simpler strategies for “weak” signals).

Why is this important? These signals are not useful enough to be overlooked by other traders and other systems. In case the bigger ones with very strong AI systems seize larger and larger portions of the market these very signals may remain the longest. One of the most efficient billion-dollar quant funds has specialized in just this approach for many years.

More complex MM strategies are practically out of reach for discrete traders, requiring too much work, beyond the capabilities of a single person or even a team.
Add to this the fact that a single position can be broken into thousands of smaller ones (to minimize adverse moves and dynamically adjust to pending and incoming orders) and this absolutely requires the use of intelligent automation.

Hiding Entries And Exits

If I wrote that brokers and market makers copy the positions of the best then many might be offended. So, I will not write this (oh wait!). This is one of the oldest tricks in the book, the simplest and very profitable, used since the beginning of the stock markets.

The best ones do it in several creative ways, from what I’ve observed in the life cycle of a fund there comes a time when entries are hidden to preserve their profitability. Too many eyes are watching, many smart people are able to “crack” the strategy and it doesn’t take intelligence at all to copy.

Some of the best go to the extent of not only hiding the inputs but also hiding the intent, the system, the strategy behind the inputs to prevent reverse-engineering it. This is understandable, if they have something that works well (for which they usually paid a lot, sometimes tens of millions) they want to use it for years.

Here I want to talk about two strategies I’ve observed, so that if someone hasn’t dealt with it yet they can form an opinion.

The first strategy is to split the inputs between two, three or more brokers. The system has a certain number of profitable inputs and a certain number of losing ones, and the split itself can cause fluctuations – perhaps more losing ones will flow to one of the brokers?

It is possible to consciously (if such a practice is suspected) direct inferior orders to any of the brokers. It is also possible to do such things routinely, changing from time to time to whom the better and worse orders are directed. This is bound to reflect on the copyists, as the system will not be as good as expected.

Another strategy requires having or running its own HFT trading department. One large fund does several hundred thousand entries a day (between two hundred and four hundred thousand entries), and among these HFT hides its entries from other systems.

This fund is operated by two very large brokers.
If someone had access to one of the brokers and tried to crack its strategies, he would first have to filter out the large-scale signals (5 minutes and above) from the tens of thousands of HFT signals. This is of course possible but … the other problem remains – neither path gets the system fully. Second – the fund uses not one strategy but dozens. And breaking this is already a non-trivial matter (but still probably possible for a patient statistician).

But then there is another matter – the fund employs a lot of excellent mathematicians who, over time, will surely realize that one path has worse results than another and ask themselves why (the fund has already had to deal with the use of its signals and knows the subject from experience).

They won’t necessarily be able to prove that someone is sneaking their signals or trying to break their strategies, but they might just want to change brokers.

This is the ending of part one. You can read the whole thing in the magazine. Just subscribe below!
STAY IN THE KNOW

Get Current [Issue] of New City Trader.
 Now It's Free!

]]> An Interview With ChatGPT About Trading https://newcitytrader.com/an-interview-with-chatgpt-about-trading/ Sat, 29 Apr 2023 11:23:29 +0000 https://newcitytrader.com/?p=8617

The ChatGPT-4 interview about trading, for simplicity’s sake, takes the form of a conversation with a live person. The development of AI will go through several stages, the first two of which are: treating AI as a tool and treating AI as a colleague. The latter we understand – treating as an assistant or colleague at work. This is the spirit in which the conversation is conducted, this is what awaits us in the near future in virtually every field of work and personal life. During the interview I repeatedly ask questions about any terms I am interested in. It is important to take into account that sometimes the answers may not be true (it happens) and in order to use the code (there is an example of the code he gave) it is not enough to copy it. To make it work you need to know a lot about the architecture of programs and programming itself.

And now the conversation…

How can ChatGPT-4 be used in preparing or making investment decisions in stock and currency markets?

ChatGPT-4, which is an advanced language model based on the GPT-4 architecture, can be used in various aspects of preparing and making investment decisions in the stock and currency markets.

  1. Sentiment analysis: ChatGPT-4 can analyze news stories, articles and social media posts to determine overall market sentiment toward specific companies, sectors or currencies, which can help investors choose the right instruments to invest in.
  2. Price forecasting: The model can be used to analyze historical price data, indicating possible trends and price directions for the future, which can be useful to investors when making decisions about short- and long-term investments.
  3. Fundamental analysis: ChatGPT-4 can help investors analyze companies’ financial reports, pulling out key indicators and information that can affect the value of a company’s stock.
  4. Identification of market events: The model can monitor news to detect significant market events, such as mergers and acquisitions, changes in interest rates or reports of geopolitical conflicts that may affect stock and currency markets.
  5. Technical analysis: ChatGPT-4 can help investors identify important support and resistance levels on price charts, which can be crucial in making investment decisions based on technical analysis.
  6. Portfolio optimization: The model can be used to evaluate the diversification of an investment portfolio, suggesting changes that can reduce risk or increase expected returns.
  7. Investment strategy suggestion: ChatGPT-4 can present various investment strategies, such as value, growth, or long-term investments, tailored to the investor’s individual preferences.
  8. Investment idea generation: The model can generate investment ideas by analyzing market data, company performance, the macroeconomic situation and many other factors.
  9. Risk evaluation: ChatGPT-4 can help the investor

Elaborate on the first point in 5 paragraphs. How can ChatGTP-4 be used?

The first point is about using ChatGPT-4 to analyze sentiment in stock and currency markets. Let’s consider this issue more broadly, presenting five aspects in which the model can be useful.

  1. identifying sources of information: Using ChatGPT-4 to analyze sentiment in stock and currency markets can begin with identifying relevant sources of information. The model can search websites, discussion forums, blogs, and social media to identify where investors and analysts are posting their opinions on companies, sectors or currencies. In this way, an investor can access a rich database of diverse and up-to-date information on financial markets.
  2. Data extraction: Next, ChatGPT-4 can be used to extract data from collected sources. The model can analyze texts, extracting key information from them, such as positive or negative opinions on companies or currencies, comments on the macroeconomic situation, or price forecasts. In addition, ChatGPT-4 can recognize emotional language in analyzed texts, which can further provide information on market sentiment.
  3. Sentiment classification: After data extraction, ChatGPT-4 can be used to classify sentiment. The model can assign sentiment ratings to analyzed information based on the language used, determining whether the opinion is positive, negative, or neutral. In this way, the investor gets a consistent picture of market sentiment that can be used to make investment decisions.
  4. Aggregation and visualization of results: Once sentiment is classified, ChatGPT-4 can help aggregate results, creating synthetic sentiment indicators for individual companies, sectors or currencies. The model can present these indicators in the form of easy-to-read charts, tables or reports, making it easier for investors to interpret the results of sentiment analysis.
  5. Use of results in investment strategies: Based on the sentiment analysis conducted by ChatGPT-4, investors can adjust their investment strategies. For example, an investor may decide to buy shares of a company with a high positive sentiment, hoping for a continuation of the upward trend. On the other hand, an investor may avoid companies with negative sentiment, which may be vulnerable to declines in value. In the case of currency markets, sentiment analysis can help investors identify currencies with strong upward prospects, as well as signal potential risks associated with currencies with negative sentiment. In this way, sentiment analysis supported by ChatGPT-4 can become an important component of both short- and long-term investment strategies.

Are you able to analyze Twitter sentiment around Apple, for example?

As an AI language model, I cannot directly search the Internet or social media in real time. My abilities are based on knowledge accumulated during training, which was completed in September 2021. Therefore, I am unable to provide up-to-date information on the sentiment around Apple on Twitter.

To analyze the sentiment around Apple on Twitter, I suggest using analytical tools dedicated to social media sentiment analysis. You can also program your own tool, using Twitter’s API and a language model such as ChatGPT-4 to analyze and classify tweets. This kind of analysis can provide up-to-date information on the sentiment around a company and help you make investment decisions.

What would it take for you to analyze sentiment around Apple on Twitter in real time?

As an AI language model, I am not able to directly perform real-time analysis. However, to enable sentiment analysis around Apple on Twitter, you can take the following steps:

  1. access Twitter’s API: To retrieve data from Twitter, you will need to register as a developer and access the service’s API. Twitter’s API will allow you to retrieve tweets related to Apple, based on keywords, hashtags or mentions of the company.
  2. Gather data: Using Twitter’s API, retrieve relevant tweets related to Apple. You can limit your search results using time, language or geolocation filters to get the most relevant data.
  3. Prepare the data: Process the collected tweets, clearing them of unnecessary elements such as links, special characters or emoticons. Make sure the data is in a suitable format that can be processed by the AI language model.
  4. Use the AI language model: Integrate ChatGPT-4 or another AI language model with the acquired data. Use the model to analyze tweets, identifying positive, negative or neutral opinions about Apple.
  5. Analyze and visualize the results: Once the AI language model has processed the data, analyze the results, aggregating sentiment at a general level and in selected categories (e.g., products, services or Apple-related events). Visualize the results in the form of charts or tables to facilitate interpretation and conclusions.

It is worth noting that real-time data analysis will require continuous monitoring and updating of input data. As a result, you may need to use appropriate streaming data processing techniques and maintain an infrastructure to support real-time analysis.

Regarding point 3, are you able to clean this data yourself?

Yes, as an AI language model, I can help clean up text data. You can use me to remove unnecessary elements, such as links, special characters or emoticons, and convert the text into a suitable format that can be processed by the AI language model during sentiment analysis. Here’s an example of how I would clean up the data:

Input: “Check out the new iPhone 13! 📱😍 https://apple.example.com #Apple #iPhone13”

Result: “Check the new iPhone 13”

However, it’s worth remembering that as an AI language model, I operate based on the text data you provide me. I am not able to retrieve tweets from Twitter or interactively process the data myself. Therefore, you will have to collect the data from Twitter yourself and then provide it to me as text so that I can help clean up the data.

Are you able to briefly explain what kind of results we can get and what valuable information about Apple we can get from them?

Sentiment analysis around Apple on Twitter can provide valuable information that can help you make investment decisions or understand the general perception of the company and its products. Here are some examples of the results we can get from such an analysis:

  1. general sentiment: The analysis can show whether the overall sentiment toward Apple is positive, negative or neutral. This can help assess the strength of the brand and understand whether investors and customers have positive expectations of the company.
  2. Product and service sentiment: The analysis can identify positive and negative opinions about specific Apple products or services, such as the iPhone, iPad or iCloud services. This can help identify the strengths and weaknesses of the company’s offerings and forecast demand for specific products.
  3. Reactions to events and announcements: Sentiment analysis can reveal reactions to specific events, such as product launches, conferences or financial results. This allows you to assess how markets and customers perceive this information and how it may affect the company’s value.
  4. Comparison with competitors: Sentiment analysis can provide information on how Apple is doing compared to competitors such as Samsung, Google and Microsoft. Sentiment comparisons can assess Apple’s position in the market and identify possible threats.
  5. Trend detection: Systematic sentiment analysis can detect changes in Apple’s perception, both positive and negative. Detecting trends allows you to react early to changing market conditions and adjust your investment strategies.

The results of sentiment analysis, combined with other market and financial data, can provide valuable information to help investors make decisions, and allow companies to better understand the perception of their operations, products and services

Let’s move on to the second point: “Price forecasting: The model can be used to analyze historical price data, indicating possible trends and price directions for the future, …” How can Chat do this?

ChatGPT-4 can be used to forecast stock and currency prices, but it is worth noting that, as a language model, its main purpose is to analyze and generate text, not numerical or statistical analysis. Nevertheless, ChatGPT-4 can be used to analyze historical price data, indicating possible trends and price directions for the future. Here’s how Chat can do this:

  1. processing historical data: The first step is to provide the model with relevant historical data on stock or currency prices. This data should include information on opening, closing, highest, lowest prices and trading volumes.
  2. Initial data analysis: ChatGPT-4 can be used to identify significant events that have affected stock or currency prices in the past. The model can search its knowledge of economic, political or technological events that may have affected prices during the period under review.
  3. Identify patterns and trends: ChatGPT-4 can be used to analyze price data to identify patterns and trends, such as ups, downs, consolidations or reversals. The model can also suggest how these patterns have affected prices in the past and how they may affect them in the future.
  4. Analytics-based forecasting: Based on analysis of historical data, ChatGPT-4 can attempt to forecast future price directions. However, it is worth remembering that these forecasts will be based on the knowledge accumulated by the model during training, and not on actual statistical or econometric calculations. Therefore, these forecasts should be treated as preliminary and complementary to more advanced methods of financial analysis.
  5. Verify and update forecasts: As new data becomes available, ChatGPT-4 can be used to update forecasts, taking into account new information. The model can also help you assess how accurate your forecasts were, identifying any errors and adjusting your forecasts for the future.

While ChatGPT-4 can be used to forecast stock prices and currencies, it is worth noting that there are more advanced methods of financial analysis, such as econometric models, technical analysis or learning algorithms you have.

And how do you physically do this? I could, for example, provide you with data on the last two hundred OHLC values for the eurodollar and based on that you can say something?

Although ChatGPT-4 can be used to analyze textual data, it is not optimized for analyzing numerical data such as OHLC data. Physically performing such analysis with ChatGPT-4 would be limited, and the results may be inaccurate. Instead, statistical models and machine learning algorithms developed specifically for financial analysis are more suitable. However, if you wanted to use ChatGPT-4 to analyze OHLC data for the eurodollar, we could try an approach based on a textual representation of the data. Here’s what that might look like.

  1. convert numerical data to text format: to enable analysis by ChatGPT-4, OHLC data must be converted to text. This can be done by presenting the data as a text table or by describing each data point with a sentence.
  2. Example: “Date: 2023-04-07, Opening: 1.1000, Highest: 1.1100, Lowest: 1.0900, Closing: 1.1050.”
  3. Provide data to the model: Then, enter the transformed text data into ChatGPT-4, giving clear instructions on OHLC data analysis and expected conclusions.
  4. Data analysis by the model: ChatGPT-4 will perform a limited analysis of the text data, identifying potential trends and patterns based on its prior knowledge of the currency markets. Note, however, that the results of such analysis may not be precise or reliable, as the model is not optimized for analyzing numerical data.

Alternatively, I would recommend using dedicated financial analysis tools, such as statistical models, technical analysis or machine learning algorithms, to get more precise predictions. For example, you can use linear regression, ARIMA, LSTM, or XGBoost models, which are better suited for time series analysis and price forecasting.

It is worth remembering that ChatGPT-4 can be useful in the context of analyzing currency markets as a support in identifying potential economic, political or technological events that may affect the Eurodollar exchange rate. This information can be useful in conjunction with more advanced financial analysis methods.

If so, are you able to create and run such an advanced model: it will give you market data (OHLC plus volume) and a list of events (based on the news of the period), What will you be able to build from such groups of information?

As an AI language model, ChatGPT-4 is not able to directly create or run advanced financial analysis or machine learning models. My main task is to analyze and generate texts, not to create and run numerical models. Nevertheless, I can help you develop a strategy for analyzing the data you provide, and point out methods that can be used to analyze this data by specialists or dedicated tools. Based on the market data provided (OHLC plus volume) and a list of events (news) from the period, here’s what you can do:

  1. conduct a technical analysis: Based on the OHLC data and volume, specialists can apply various technical analysis methods such as indicators (e.g., RSI, MACD, moving averages) and chart formation analysis (e.g., triangles, head and shoulders) to forecast future price movements.
  2. Conduct fundamental analysis: A list of events and news stories can be used for fundamental analysis to assess the impact of the news on the value of a stock or currency. Fundamental analysis can include an assessment of economic, political and technological conditions and their impact on the value of the instrument being analyzed.
  3. Use of statistical models and machine learning: Based on market data and event information, various statistical and machine learning models such as linear regression, ARIMA models, neural networks (e.g., LSTM) or XGBoost can be used to forecast future price movements. News list events can be included as explanatory variables in these models.
  4. Integrate different analysis methods: Technical analysis, fundamental analysis, and statistical and machine learning models can be combined for more accurate forecasts. Such integrated analysis can take into account both historical information from market data and the impact of economic, political or technological events on the value of a financial instrument.
  5. It is worth remembering, however, that as ChatGPT-4, I cannot directly create or run such models. My role is to provide knowledge and guidance on how these methods can be applied to data analysis. In fact, you will need financial analysis specialists.

But if I wanted you to write me code in Python to build such a model are you able to do it?

Yes, I can help you write basic code in Python that can serve as a starting point for further development and customization. Here is an example of a simple predictive model based on linear regression that can be used to analyze OHLC data:

python
import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error

# Loading the data (assuming the data is stored in a CSV file)
data = pd.read_csv("your_data.csv")

# Pre-processing the data
data["Date"] = pd.to_datetime(data["Date"])
data.set_index("Date", inplace=True)

# Create features (features) and target variable (target)
X = data.drop("Close", axis=1) # Remove the "Close" column to make the other columns serve as features
y = data["Close"] # Use the "Close" column as the target variable

# Divide the data into training and test sets

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# Create and teach a linear regression model
model = LinearRegression()
model.fit(X_train, y_train)

# Predicting on the test set
y_pred = model.predict(X_test)

# Calculation of the root mean square error (RMSE).
rmse = np.sqrt(mean_squared_error(y_test, y_pred))
print("RMSE:", rmse)

The code above reads data from a CSV file, processes it, creates features and a target variable, and then divides the data into training and test sets. It then creates and teaches a linear regression model, predicts values on the test set, and calculates the root mean square error (RMSE) as a measure of model quality.

It is worth noting that this code serves as an example of a simple model and may require further customization and optimization. For more advanced models, such as neural networks (LSTM) or XGBoost, the code will be much more complex. In such cases, I recommend consulting data analysis or machine learning specialists for a more advanced solution.

This is the ending of part one. You can read the whole thing in the magazine. Just subscribe below!
STAY IN THE KNOW

Get Current [Issue] of New City Trader.
 Now It's Free!

]]> Decision making process in trading is one of the most important concepts. But WHY? https://newcitytrader.com/decision-making-process-in-trading/ Mon, 07 Mar 2022 08:00:00 +0000 https://newcitytrader.com/?p=6633

Trader’s decision making process is one of the most important concepts in trading for me. Why?

Results depend on a good process.

What is the decision-making process?

Generally, it’s all the thinking and analytical processes and the decisions you make based on them, from the initial analysis (“idea”) to closing and analyzing a position.

The entire thought process that goes on in your head.

This approach came to my mind, even imposed itself when I was wondering how AI will make decisions – the best possible ones.

Then the same principles were already obvious in application to traders. And even later it turned out that the best ones have specific elements of decision-making process practically absent in other traders.

So much history, now what came out of it? Let’s break down your decision-making process into elements.

1. It starts with analyzing the markets and choosing an instrument (specific stocks, specific currencies from the available pool – for example)
What factors drive your choice, why did you choose this instrument and not another?

Here we have the first decisions. On what basis do you make them?

This is important, because if you start to think about how to improve your decision-making process in order to have better results – you will realize that … maybe you can choose better.

Let me tell you how the best do it: The best learn to choose the best inputs from those available to them. In the sense that if they specialize in a sector of companies, they choose the shorts or longs with the best prospects for profits.

The best traders try to pick the best of the many opportunities that are available to them.

The best traders try to pick the best opportunities from among the many that are available in the markets they trade.

What does that mean for you?

It means that it is possible and even worthwhile to work on the criteria of selecting the instrument on which it will be the easiest to choose, on which we will earn the most (potentially).

The criteria may be different for different traders, but the idea is the same:
You can choose the instrument, the market situation that is best for you.
In the case of AI systems, their development will initially go in this direction: searching for the best opportunities from what is available. It will take, for example, 0.5 seconds to search five thousand companies in this respect. And from this enormity it will choose e.g. 100 for further analysis and playing.

I am mentioning this to start slowly realize what kind of analytical and decision-making power we will have to deal with.

But let’s go back to the decision-making process.

2. We have selected pools of top companies, what next? Let’s assume that we want to enter as long as the criteria are right.

And here is another question: where, among the selected are the best situations?

A simple example: we want to play longs on companies with good fundamentals.
Our selection factor at this stage of the decision-making process will be (for example) volume behavior.

The idea is to enter such movements, where the volume is clearly increasing – which suggests that the interest of buyers is growing.

At this stage we will rank our companies according to the best situations of increasing volume.

Somewhere it is too low, somewhere it is too high and the price is not growing much (which can be suspicious, because it means that someone is buying a lot and someone else is selling, if in a good technical situation someone is selling it cannot be a beginner, it is rather a professional, which can mean that they know more than us and get rid of the stock).

At this stage we have a few companies with the best situation.

Here we used volume, but we can analyze more elements: candlestick patterns, Fibonacci levels, MACD, CCI and thousands of other indicators.

How about using the confluence of indicators and situations?

For example, the price reaches the average 200 from the top, is close to resistance and at the Fibonacci level of 62%. Here you meet several things at once that other traders are looking at.

I want to emphasize that at this stage of the decision-making process we also have several options to choose from.

Let’s ask ourselves which indicators are best for our purposes?

It may not be the volume, but for example the relation between the Bid and Ask order volume.

The Bid is falling – the number of sellers is decreasing, which may mean that the market is slowing down. After a while the expected situation occurs – Bid orders appear, large orders.

I have seen situations where traders enter both the expiring Bid (upcoming ASKs are a signal to confirm the positions, when there are no ASKs the trader runs away from the position) and the upcoming large Ask orders.

To summarize: we can choose the best market situation according to many indicators, many criteria.

3.

Now it’s time to decide how much capital we will put on which one. Let’s stop our discussion for a moment and summarize the important points:

To summarize the knowledge so far:

The decision-making process is a series of analyses and decisions.

Each distinct element of the decision-making process can be improved, approached from different angles, analyzed and refined according to different criteria.

This gives you the opportunity to improve the results – by improving each element on which the results depend.

A recipe for performance improvement:

Break down the way you make decisions into elements.

Learn as much as you can about how you can improve each of them.

Of course, there’s more, but I don’t want to write too much here, so it’s worth leaving traders with this interesting idea – pointing out the process that will improve their results.

Someday I’ll write more about the stages and best practices that can be used there.

And about the fact that the psyche plays a huge role in this process (which cannot be seen from my rational and mechanical description).

Let me give just one example:

Getting to TP – due to the pressure of the markets traders often run away from positions, emotions, pressure on the psyche is too strong. The psyche is the filter through which all decisions pass. And it can disrupt each decision and each step – from the initial analysis to the execution of the entry.

These are some topics for our future meetings.

Emotion as a key element in market decision making

The decision-making process consists of two components: rational (analytical) and emotional.

The more experienced a trader is, the more importance he attaches to the emotional component – but in a specific sense.

This does not mean that emotions dictate what to do.

Very experienced traders treat their emotional state as an additional source of information, and can stand by and hear the “gut feel” above and beyond the emotional noise.

To quote one experienced trader:

“If I am feeling the pressure in this situation it means that others in the market are also feeling it. The one who can withstand it better will win.”

In the same situation, inexperienced traders may unwaveringly believe in the high quality of decisions made on the basis of “nose”.  That’s how one retail trader I know lost more than €750,000 in a few months.

The best traders use various techniques to objectify their decisions – knowing that emotions can and will get in the way of their sober assessment of the developing situation and their ability to make good decisions.

According to a study of institutional traders – in the same situation less experienced traders use strategies to avoid emotions.

This makes a difference and a barrier for beginner and intermediate traders. The best traders are able to withstand large, sustained pressure by entering into it consciously, while the less experienced traders will at some point succumb and run away from the same position.

Both groups of traders differ not only in their experience, but also in their strategies of dealing with emotions. Because it turns out that experience is not the only thing that matters. Research shows that traders with a lot of experience (e.g. more than 20 years) who do not use the same strategies for dealing with their emotions as the best traders do, tend to underperform.

The way of dealing with emotions is a clear differentiator for the best traders. It allows you to survive chaos, volatility, uncertainty and panic, it allows you to continue to enter the market when others have long given up, and it opens new doors to results.

We are preparing quite a training on this topic of how the best deal with emotions.

Contact us if you are interested: magazine[@]NewCityTrader.com

STAY IN THE KNOW

Get Current [Issue] Now It's Free!

]]> How traders can compete with computers? Can we? https://newcitytrader.com/how-traders-can-compete-with-computers/ Mon, 28 Feb 2022 08:00:00 +0000 https://newcitytrader.com/?p=6599

The question is how much traders outside of the large funds will be able to compete with a computer that has the decision-making process of the best traders encoded and places positions in hundreds of markets simultaneously I leave open. I don’t think it’s that hard to answer.

We are maybe 15-20 years away from that point (the deadline for a full SUPER AI solution in trading) but already within the next 6-8 years I think its main components operating in and perhaps dominating the most profitable, liquid markets will be created.

In this short period of time (maybe it will be even shorter) the strongest AI solutions will have a chance to dominate the market and will only be limited by government regulations (if any are created, which is not certain).

In short, in maximum 15 years AI will dominate the markets and the whole investment market in general will be divided between two, maximum three big solutions.

STAY IN THE KNOW

Get Current [Issue] Now It's Free!

]]> The future of Alternative Intelligence in trading https://newcitytrader.com/the-future-of-alternative-intelligence-in-trading/ Mon, 21 Feb 2022 08:00:00 +0000 https://newcitytrader.com/?p=6598

AI solutions will continue to mature. They will become more efficient and faster. At the beginning they will surely support us in various ways, providing knowledge, tips, automating tasks and processes.

This will be the time when the best results will be achieved by mixed teams (I would call them hybrid teams) composed of traders and AI.

An article on this topic is already in preparation, I will show in it the right direction of development, the next step of AI solutions adoption by funds. So far it is not well understood and I want to change that.

Ultimately, Super AI will be created, which will collect data on its own, analyze them and make investment decisions. I will anticipate a possible question – the AI I have in mind will work like an excellent trader, the best of the best in the field.

And the cost (variable) of creating and testing the new system will be equal to the price of electricity used by the computer.

The time to build a valuable strategy for using AI in funds is today.

The next important step is not to bet on a “standalone AI machine” (as many fund managers believe) but on a hybrid team of well-prepared and trained humans (including necessarily traders) and evolving AI.

STAY IN THE KNOW

Get Current [Issue] Now It's Free!

]]> It is Alternative Data that gives traders and funds the biggest advantage today, not AI https://newcitytrader.com/alternative-data-biggest-advantage/ Mon, 07 Feb 2022 12:09:57 +0000 https://newcitytrader.com/?p=6589

A fierce battle is underway for better returns, reduced risk and faster entries. It is Alternative Data that gives traders and funds the biggest advantage today, not AI. The time of AI dominance is yet to come.

Alternative Data is a revolution in trading today more important than the Artificial Intelligence revolution. They are able to improve virtually every part of a fund’s investment activity: improve signals or systems, reduce risk, show situational context when trading a position.

The system hasn’t stopped working, but there’s no liquidity like there used to be, entries cause stronger counter moves because… the biggest ones have already been there and exhausted the liquidity. The biggest funds use alternative data to enter faster.

Better pre-trade analysis improves performance. Better post-trade analysis improves understanding of the factors that make up good performance. Until now, such deep insights into the market and the environment have been impossible.

From a strategy perspective, funds have no choice today and (sooner rather than later) dealing with alternative data will be a necessity. From an operational strategy point of view – it is important today to create in the fund the competence to use alternative data, because this revolution is just beginning.

Alternative data gives a big market advantage. Today, in addition to AI tools, they are a major driver of building competitive advantage. However, what they bring to some they take away from others. That means systems stop working, signals become less profitable, pressure on discretionary traders and portfolio managers increases. AD are changing markets without a doubt.

In a recent article in New City Trader I show more broadly, with specific examples, how they are doing this ⇓

STAY IN THE KNOW

Get Current [Issue] Now It's Free!

]]> Alternative Data is a revolution in trading. Today even more important than Artificial Intelligence https://newcitytrader.com/alternative-data-is-a-revolution-in-trading/ Thu, 02 Sep 2021 07:04:10 +0000 https://wgl-demo.net/thepascal/?p=210

The use of Alternative Data is the current super trend strongly shaping the present and future of investing. It is a revolution more important than the revolution that Artificial Intelligence will give.

The main benefit of Alternative Data (AD) is that you can have most of your important metrics faster and more accurately than ever before.

Alternative Data currently improves signal quality and reduces risk. In the future, it will be the primary source of signals. Today, it is the main source of competitive advantage in many cases to find and use these signals before others.

In addition to the current use of AD, it is imperative to create competencies in the fund to use it wisely in the future because this area will now evolve rapidly.

Thanks to AD, the signal will be received earlier. As a result, it will be better, less risky, the position will be easier to run, and the exit will be better.

I have prepared 20 examples of using alternative data. I want to show a broad spectrum of situations where AD have been used so far and can be used in the future.

It is often the case that an exciting and inspiring idea comes from a completely unexpected direction. Therefore, it is worth attending industry conferences and collecting as many case studies as possible of what others have done.

Alternative Data New City Trader

Counting cars in the Tesla parking lot

One fund applied Machine Learning (ML) tools to analyze satellite images of the parking lot in front of Tesla’s Megafactory. The tool analyzed the position of cars and their colours. The goal was to determine if and how the number of vehicles was changing. It turned out that cars quickly disappear and new ones appear in their place, which indicated that the company would keep its commitments and plans. Moreover, with this analysis, it was known weeks before the official announcement. Such information provides a substantial advantage over other market participants.

The above is a rather famous and awe-inspiring example, so it is worth commenting on it. If we look at the area around this factory (in Nevada), we can see one access road.

Setting up a car with a camera there and having someone count the vehicles leaving the parking lot would give the same result (and perhaps much cheaper) as using a whole team to machine-analyze satellite images. It is worth knowing this and looking for a way to have the same data not for a quarter of a million dollars but for 1% of that amount.

Non-Farm Payrolls Employment Data...

Full version of the article is available for you in New City Trader magazine (now free of charge):

STAY IN THE KNOW

Get Current [Issue] Now It's Free!

]]> Artificial Intelligence in trading – 6 hottest topics https://newcitytrader.com/artificial-intelligence-in-trading/ Wed, 01 Sep 2021 07:02:29 +0000 https://wgl-demo.net/thepascal/?p=218

Inthis article, you’ll learn about six of the most critical and “hottest” elements that make up or are associated with Artificial Intelligence in trading today. As an addition, you’ll learn the basics of the tools that will shape the future of trading. First in the biggest and wealthiest funds and then in smaller ones as well.

I invite you to take a journey into this near and far future of trading.

NLP

Natural Language Processing is the common name for many tools to analyze written and spoken text: company documents, press articles, news, analysis, web pages, social media posts, company’s product reviews.

Advanced NLP software recognizes context up to about a thousand words. That’s a lot, and soon, there will be more.

The NLP allows you to analyze many features of text, such as whether:

– the text about a particular company is positive or negative,
– it is clear and transparent or obscure and convoluted,
– the authors express themselves positively or negatively about the future.

When we analyze texts of reports and press statements, it turns out that all of the above elements can be a good indicator of future financial performance

Social media texts

Already now, the analysis of posts in social media and online shops allows determining the sentiment – the opinion about the company and its products, which usually precedes the financial results.

Sometimes it is also possible to find and analyze the sentiment of different investor groups about the company and its future which also affects the share price.

"The stripper that will change our World"

The next gigantic step in the evolution of NLP come from extracting written knowledge from millions of books, academic articles and other texts and help create coherent theories of how the economy, or supply chains, works.

This development will give theoretical and practical insights into the factors that affect the financial performance of companies, industries and all relevant economic processes. Already today, we see the first signs of the creation of such tools.

Spoken text

NLP also covers spoken word analysis: statements from TV news, films, other video material, or telephone conversations are automatically transcribed and subject to the same analysis as written content.

Thanks to this, we now have access to knowledge about the level of Forex volume. The systems analyze the volume at major banks and brokers and traders’ volume over the phone. Until last year, data on this was not available in real-time. I will find out if anything has changed yet and write about it in future issues. I know that there were plans to provide this volume in real-time as well.

Full version of the article is available for you in New City Trader magazine (now free of charge):

STAY IN THE KNOW

Get Current [Issue] Now It's Free!

]]> [Full] Artificial Intelligence: The lowest-hanging fruit. Part I https://newcitytrader.com/artificial-intelligence-in-trading-lowesthanging-fruit/ Wed, 01 Sep 2021 07:00:16 +0000 https://wgl-demo.net/thepascal/?p=233

I recently spoke with an elderly gentleman – a trader and fund owner. This conversation inspired me to write an article about Artificial Intelligence tools that are used in trading today. His fund employs over a dozen traders investing in various markets, and he is a veteran of oil trading. His trading style is conservative – after finding a signal, he opens a trade holding a single position, sometimes for several weeks.

I want to show you where and how you can use the most modern solutions in this example. The largest funds use such solutions daily.
As we know, a trader’s performance depends on his decision-making process, signal quality, available liquidity and what happens during position management. New Artificial Intelligence tools will improve each stage of this process.

The solutions and ideas can be applied to any discrete trading system.

Alternative data means more accurate price forecasts and better, less risky signals

Many years ago, to find out the financial results of a chain of stores, one had to wait for their earnings announcement. Today, the analysis of satellite images of parking lots makes it possible to estimate whether there are more or fewer customers and, on this basis, to forecast the results.

Data from credit card transactions also allow estimating whether there will be an increase or a decrease in sales in the chain.
By analyzing card transactions and various data from the company’s environment, we can estimate the financial results that are yet to be published by this company. If we have access to relevant data, we can make such estimates ongoing and make investment decisions.

For the oil market, some companies forecast price movements based on inventories and demand. Today, you can count the amount of oil in tanks in remote areas of China by measuring the length of the shadow they cast (using satellite images). To this estimate, the amount of oil stored in tankers and on railway sidings is added. Thanks to this, we know the size of stocks in virtually every country “right now”.

Today’s “forecasting” goes to the real-time estimation of the selected quantity. It is one of the strong trends changing investing. As a result, traders and investors have a better picture of the current market situation. Consequently, entries are more precise and less risky. And funds that build this kind of competency at home are winning in the markets.

Supertrend: Moving from forecasting to nowcasting.

Traders who traded moves after the news are now seeing that these moves are getting weaker. On the other hand, traders in the largest mutual funds are rarely surprised when publishing data because they can estimate it in advance. It is also one of the reasons for the disappearance of some systems.

I read an article a few months ago by an economics professor arguing that alternative data is unnecessary because large funds react to earnings releases anyway. Of course, there is some point in this. However, suppose someone can assess the data before publication. In that case, he has a chance to enter the market faster – even before the others. Or take advantage of the entry of other large investors to exit the market with profits at the right time.

Market behaviour is changing because major investors have changed their behaviour. Those who fail to recognize this in time will be left with old signals and many stocks that are no longer as attractive as before.

Entry and exit - minimizing costs and adverse moves

One of the main problems that most institutional traders complain about today is the difficulty in finding liquidity.

The automated tool can analyze the available liquidity with different brokers, plan and execute market entry. Their use is to take advantage of available liquidity with minimal cost and minimal adverse move.

Virtually every large fund has this type of tool. For example, quants analyze order structures using Machine Learning methods and, on this basis, program various entry strategies. As a result, the program will execute the entry faster and better than a large team of traders.

Automation gives the trader more time for other tasks – more important than position building.

Trend: Entry automation: The program will execute an entry faster and better than a large trading team. The trader has more time for other tasks – more important than position building.

Entry when the signal is very good

Low risk attracts significant capital. Moreover, new signals are getting better and less risky, so the automated trading solutions get new tasks: increasing positions to the maximum for the best signals. 

This change makes it difficult for other traders to find liquidity. As a result, they cannot enter the market as before, and the adverse movement is more substantial than previously.

However, this is not a problem for smaller traders. Some consider the lack of liquidity to confirm high signal quality. To quote:

“Lack of liquidity in good companies means that the biggest ones are already there.”

This situation is a problem for large traders who do not use the latest tools. The signals are still good and followed by the expected moves. However, there are no longer opportunities to build large positions.

The struggle for liquidity, especially in the best market situations, will only increase. Therefore, the focus will be on getting a good signal as early as possible to enter with significant capital quickly. It is with alternative data that this becomes possible.

The next step will be to exploit weak signals (I will describe what they are and how to exploit them in a future article).

Better signals and lower risk attract more capital. Automation of entries and exits becomes necessary; the program can utilize the available liquidity to the maximum. The competitive battle focuses on having a good signal as soon as possible.

Better assessment (view) of the situation while taking positions

While taking positions, a better view of the situation comes with new tools. However, this improvement does not end with market entry. Social media sentiment analysis and automated news analysis gives you a picture of market interest in a particular instrument.

Example:

One of the funds uses the mean reversion system. If the price of a company after good news has moved too far away from the average, then the program places a sell order.

The basis of this system is pure statistics. The program counts the main moving average the deviation from it. The system monitors the price, and if the deviation is “too high”, it places a sell order back to the average.

However, when good news keeps coming, the market often continues moving up.

To cope with this situation, they added a module for analyzing news and sentiment in social media. Suppose the information about the company is still good (or there are many positive posts on social media). In that case, the sell position is automatically closed.

Profit increase achieved.

Discretionary traders can use the same scheme. However, while it is still possible to follow the news for a few positions, it is difficult to follow and create any meaningful picture of social media sentiment. The Machine Learning tools for analyzing Twitter, Reddit, Finwiz, FB or any media where potential traders can undoubtedly help here.

Knowing what others think about a stock or commodity can help you manage your position better. If the sentiment gets bad, we can exit the market earlier, and a good social mood can make us hold the position.

Of course, these tools should be approached with caution – the market is dominated by the most significant players, and they do not publicly announce their intentions. But they have these tools, and with their aid, the smaller players can see what the big ones see and thus better understand their intentions and decisions.

Today, you can shorten or lengthen a position with news and social moods analysis because you have a better picture of what others are seeing, reading, and thinking.

Availability of the new solutions

The tools described here are available to the largest funds. However, their development is expensive and requires teams of good specialists (machine learning, social media analysis, liquidity analysis, microstructure of markets and its evolution over time).

As time goes on, these solutions will become more widespread, available from larger brokers or offered as stand-alone solutions.

Conclusion:

Thanks to alternative data, we see a shift from forecasting to nowcasting. We also have better forecasts of supply, demand and price behaviour. As a result, trading signals are better and less risky.

Current tools allow us to make better use of available liquidity, minimize entry costs and get a better average price. Over time, these two goals will be joined by a third: an emphasis on exploiting the best signals to the maximum.

Less risk will attract more capital, and liquidity will be depleted in many excellent current oves over time. The struggle for liquidity will intensify further. As a result, many sound systems will continue to yield high-quality signals. However, there will be no way to build a bigger position. Systems will still be good, but they will no longer be as profitable as before.

There will be a struggle to improve signal quality in the long run, mainly to make the signal appear faster.

These changes may, in some time, hit the best traders, like my interlocutor – a trader with enormous market experience. Why exactly them? People in his group have the most experience and the best signals. These are the signals that the biggest traders/funds will hunt for and try to improve.

This evolution does not mean that new tools are readily available or cheap. On the contrary, creating them requires a team of good Machine Learning specialists in finance. It also entails gathering much knowledge about the microstructure of markets.

In the next issue:

– The fastest path to cash in a fund – what to support with new tools first – some practical tips.
– AI and entry and exit hiding.
– Evolution of tools towards dynamic MM.
– Position and environment analysis.
– New and improved efficiency analysis of market paths.
– Automatic portfolio optimization (including new optimization models).
– New models for managing the size of positions on the horizon.

Full version of the article is available for you in New City Trader magazine (now free of charge):

If you enjoyed this insight:

Support Trading Education With Cake and Coffee ↷

Donate via PayPal
STAY IN THE KNOW

Get Current [Issue] Now It's Free!

]]>