hi, great video. For better accuracy you can try adding different columns: pct_change would be great. Also you should not use just Open Close High Low, they have very small difference (thats why so correlated)
For anyone wanting this code, I put a link in my description of this video to the code. Copy and paste into your Jupyter notebook. Also, I put the dataset used. Save it as a csv file.
Unfortunately this is not correct. The model has a look ahead bias, you are not supposed to know the volume, high and low until the end of the day. So basically you are trying to predict something you already know. This explains why the model has good performances. Don't think that finance is this easy guys
Yes it is not easy! Predicting stock prices is extremely challenging due to the inherent complexity and volatility of financial markets. Stock prices are influenced by a vast array of factors, including company performance, market sentiment, economic indicators, geopolitical events, and even unpredictable phenomena like natural disasters or technological breakthroughs. Moreover, stock prices often reflect collective human behavior, which can be irrational and difficult to model. Even with advanced machine learning algorithms and historical data, the dynamic and often non-linear nature of the market makes it nearly impossible to predict prices with high accuracy consistently. As a result, stock price predictions should always be approached with caution and viewed as estimates rather than certainties.
Good video :) Fun topic I thought about this topic the other day, you went about it in the best way you could. Thanks for the video
Thank you. :-) I am happy you liked it.
hi, great video. For better accuracy you can try adding different columns: pct_change would be great. Also you should not use just Open Close High Low, they have very small difference (thats why so correlated)
Thanks so much for sharing.
Please make more videos on similar topic stock prediction/training if possible
For anyone wanting this code, I put a link in my description of this video to the code. Copy and paste into your Jupyter notebook. Also, I put the dataset used. Save it as a csv file.
Is it possible to put a link to the datasets and the code because I find it difficult to write the code from the phone?
Hello. Here is the link to the dataset. Hope this helps. www.kaggle.com/datasets/shreenidhihipparagi/google-stock-prediction
Hello, I put a link in my description of this video to the code. You can copy and paste into your Jupyter notebook.
Would you please provide me the dataset?
Hello. Here is the link to the dataset. Hope this helps. www.kaggle.com/datasets/shreenidhihipparagi/google-stock-prediction
I want this code
I put a link in my description of this video to the code. You can copy and paste into your Jupyter notebook.
is possible to have the code?
Hello, yes, of course. I put a link in my description of this video to the code. You can copy and paste into your Jupyter notebook.
Unfortunately this is not correct. The model has a look ahead bias, you are not supposed to know the volume, high and low until the end of the day. So basically you are trying to predict something you already know. This explains why the model has good performances. Don't think that finance is this easy guys
Yes it is not easy! Predicting stock prices is extremely challenging due to the inherent complexity and volatility of financial markets. Stock prices are influenced by a vast array of factors, including company performance, market sentiment, economic indicators, geopolitical events, and even unpredictable phenomena like natural disasters or technological breakthroughs. Moreover, stock prices often reflect collective human behavior, which can be irrational and difficult to model. Even with advanced machine learning algorithms and historical data, the dynamic and often non-linear nature of the market makes it nearly impossible to predict prices with high accuracy consistently. As a result, stock price predictions should always be approached with caution and viewed as estimates rather than certainties.
Thank you ChatGpt
@@lefo4326 No Problem. Thank you for your comment.
Terrible sound. Can`t watch.
Sorry to hear that.