Stock Market Predicttion

Is it possible to predict stock market with zero code without using any kernels like python kernel?
Can we use timeseries/LSTM to get the prediction for the future date?

Let me formulate it this way:

It is possible to to make a model that will generate and output of what the model “thinks” will happen. But this will have nothing to do with reality and what actually will happen with the stock market.

Now predicting the Stock market is a science itself for obvious reasons (much, much money to be made). Such predictions might work in the very,very short time intervals. See high-frequency trading. Note that you actually pay a very hefty fee to get your server placed as close as possible to reduce network latency. We are talking about micro-seconds mattering and obviously highly specialized algorithms finely tuned and improved over decades. No way you can compete in any way. 0 chance.

On top of that the main aspect of a function free market is, that it by definition is unpredictable. See all your perfect algorithms would have completely failed you earlier this year because there was no chance they could predict and factor in COVID-19. Or let’s imagine an Earth quake, the so called big one, hits California. Or even a single company doing some serious lying and fraud (see Theranos) and you invested a lot into that company indirectly. The stock market is actual not rational and investors actually act mostly emotionally (else Theranos would have never happened).

If you think you can make money, forget it.
If it’s just a hobby project and you will NOT act on the predictions, well go on and have some fun.

1 Like

This topic was automatically closed 182 days after the last reply. New replies are no longer allowed.