Discussing the article: "Analyzing all price movement options on the IBM quantum computer"
This is an interesting article, however, I would like to criticize it.
- The SHA-256 encoding is quite improper choice here, because
- Cryptographic hashes are explicitly designed so that small changes in the input produce pseudo-random, uncorrelated outputs.
- Using a SHA-256 hash as your feature representation is like saying: “First, I carefully destroy all structure in my data, and then I analyze the pseudo-random bits and look for patterns.”!
- Parameter tuning is weak! You explicitly say constants like a = 70000000 and N = 17000000 were picked empirically to work optimally with financial time series.
- But you don’t show:
- How did you chose them and over what time period?
- Whether you used a separate holdout set?
- Whether you tried many combinations and then only reported the best-looking ones?
- Everything runs on a simulator, not a real IBM quantum device. This matters because:
- Simulators are just classical programs; any speedup claims are irrelevant unless you compare with an equally optimized classical algorithm.
- Real hardware noise and limited coherence would further degrade any already-weak signal.
Well, the way in which your algorythm is coded, shows flaws and it s wrong at several levels
1) Always the prediction is "0" BEARISH , whatever is the symbol or the timeframe used
2) regarding the SHA256, you should read what said my collegue. the idea at the begin, sounds amazing , but it's not properly used here
3) there is a mistake in your code
Instead of
rates = mt5.copy_rates_from_pos(symbol, timeframe, n_candles, offset )
put => rates = mt5.copy_rates_from_pos(symbol, timeframe, offset, n_candles)
If you think I am just a beginner,
take a look at this webpage => https://www.mql5.com/en/docs/python_metatrader5/mt5copyratesfrompos_py
So, please, correct the provided code
Rgds
- www.mql5.com
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
Check out the new article: Analyzing all price movement options on the IBM quantum computer.
While most traders still rely on classic indicators and patterns, quantum computers open up completely new horizons for us. With the help of the Qiskit library and quantum computer from IBM, we can look beyond conventional technical analysis and explore the market at a quantum level, where every possible price movement exists in a state of superposition.
But let's put aside the loud statements and look at the facts. Quantum computing is not a magic wand that solves all trading problems. It is a powerful tool that requires a deep understanding of both financial markets and quantum mechanics. And this is where the fun begins.
In this article, we will look at the practical implementation of quantum market analysis using a combination of MetaTrader 5 and Qiskit. We will create a system capable of analyzing historical data through the prism of quantum states and attempt to look beyond the market event horizon. Our approach combines classical probability theory, quantum phase estimation (QPE), and modern machine learning methods.
Why did this become possible now? First, quantum computers have reached a level of development where they can be used to solve practical problems. Second, libraries like Qiskit have emerged that make quantum computing accessible to ordinary developers. And third, we have learned to effectively transform financial data into quantum states.
Our experiment began with a simple question: can we use quantum superposition to simultaneously analyze all possible price paths? The answer turned out to be so intriguing that it turned into a full-fledged study, which I want to share with the MQL5 community.
Author: Yevgeniy Koshtenko