I have the following setup for a Finance/Machine Learning research project at my university: I'm applying a (Deep) Neural Network (MLP) with the following structure in Keras/Theano to distinguish outperforming stocks (label 1) from underperforming stocks (label 0). In the first place I just use actual and histrorical valuation multiples. Because it is stock data, one can expect to have very noisy data. Furthermore a stable out of sample accuracy of above 52% could be already considered as good in this domain.

The structure of the network:

  • Dense Layer with 30 features as input
  • Relu-Activation
  • Batch Normalization Layer (Without that, the network is partly not converging at all)
  • Optional Dropout Layer
  • Dense
  • Relu
  • Batch
  • Dropout
  • ....Further layers, with the same structure
  • Dense Layer with Sigmoid activation

Optimizer: RMSprop

Loss-Function: Binary Cross-Entropy

The only thing I do for pre processing is a rescaling of the features to the [0,1] range.

Now I'm running into a typical overfitting/underfitting problem, which I would normally tackle with Dropout or/and L1 and L2 kernel regularization. But in this case both Dropout and L1 and L2 regularization have a bad impact on the performance, as you can see in the following charts.

My basic setup is: 5 Layer NN (incl. input and output layer), 60 Neurons per Layer, Learning Rate of 0.02, no L1/L2 and no dropout, 100 Epochs, Batch Normalization, Batch-Size 1000. Everything is trained on 76000 input samples (nearly balanced classes 45%/55%) and applied to approximately the same amount of test samples. For the charts I have only changed one parameter at a time. "Perf-Diff" means the average stock performance difference of stocks classified as 1 and stocks classified as 0, which is basically the core metric at the end. (Higher is better)

Regularization In the l1 case the network is basically classifying every sample to one class. The spike is occuring because the network is doing this again but classifies 25 samples randomly correct. So this spike should not be interpreted as a good result, but an outlier.

The other parameters have the following impact: enter image description here

Do you have any ideas how I could improve my results? Is there an obvious error I'm doing or is there an easy answer to the regularization results? Would you suggest to do any kind of feature selection before Training (e.g. PCA)?

Edit: Further Parameters: enter image description here

  • 2
    $\begingroup$ Just some unorganized thoughts: Your batch size is pretty large. Another regularization strategy is max norm: cap the total L2 norm of the weights at a particular layer at some value. Stock data naturally has a time-series nature, so maybe an RNN layer could help. But configuring an ANN is more art than science -- even when a problem is simple, it takes the right combination of parameters for the network to learn. $\endgroup$
    – Sycorax
    Mar 20, 2017 at 20:55
  • $\begingroup$ @Sycorax: I just edited my question and included Max-norm and Batch size: Interesting behavior, but no big effect on the test results, more on the train results. Unfortunately I do not really have timeseries data but more point in time snapshot data, so I think an RNN is not applicable in my case. $\endgroup$
    – SebastianB
    Mar 21, 2017 at 9:50

1 Answer 1


Given that it's financial data, it's likely that the feature distributions in your train and validation sets are different - a phenomenon known as covariate shift - and neural networks don't tend to play well with this. Having different feature distributions can cause overfitting even if the network is relatively small.

Given that l1 and l2 don't help things I suspect other standard regularization measures like adding noise to inputs/weights/gradients probably won't help, but it might be worth a try.

I would be tempted to try a classification algorithm which is less affected by the absolute magnitudes of features, such a gradient boosted treees.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.