Vanishing when using the price difference - machine-learning

Question

With a team, we are studying how it is possible to predict the price movement with high-frequency. Instead of predicting the price directly, we have decided to try predicting price difference as well as the features. In other words, at time t+1, we predict the price difference and the features for time t+2. We use the predicted features from time t+1 to predict the price at time t+2.
We got very excited, because we thought getting good results with the following graph
We got problems in production and we wasn't known the problem till we plot the price difference.
Here is the content of the config file
{
"data": {
"sequence_length":30,
"train_test_split": 0.85,
"normalise": false,
"num_steps": 5
},
"training": {
"epochs":200,
"batch_size": 64
},
"model": {
"loss": "mse",
"optimizer": "adam",
"layers": [
{
"type": "lstm",
"neurons": 51,
"input_timesteps": 30,
"input_dim": 101,
"return_seq": true,
"activation": "relu"
},
{
"type": "dropout",
"rate": 0.1
},
{
"type": "lstm",
"neurons": 51,
"activation": "relu",
"return_seq": false
},
{
"type": "dropout",
"rate": 0.1
},
{
"type": "dense",
"neurons": 101,
"activation": "relu"
},
{
"type": "dense",
"neurons": 101,
"activation": "linear"
}
]
}
}
How can we fix the problem of the vanishing price difference?

Answer 1

Predicting the next price to within 1% or so is very easy. You could just use the old price as a prediction and achieve 1%. Prices don't change very fast. Therefore, the next price is almost always very close to the last price. The model has picked up on that.
I guess your model learned almost nothing except the very simple relationship that the next price is close to the last price. There is not necessarily anything wrong with your model. Predicting stock prices should be a very hard problem. The solution to this problem would make you rich. It is that hard.