How to make a distinction between item feature and environment feature?

Question

My data is stock data with features such as stocks' closing prices.I am curious to know if I can put the economy feature such as 'national interest rate' or 'unemployment rate' besides each stocks' features.

Data:

  Date  Ticker  Open  High  Low  Close  Interest  Unemp. 
  1/1    AMZN    75    78     73   76     0.015     0.03
  1/2    AMZN    76    77     72   72     0.016     0.03
  1/3    AMZN    72    78     76   77     0.013     0.03
  ...    ...     ...   ...    ...  ...    ...       ...
  1/1    AAPL    104   105    102  102    0.015     0.03
  1/2    AAPL    102   107    104  105    0.016     0.03
  1/3    AAPL    105   115    110  111    0.013     0.03
  ...    ...     ...   ...    ...  ...    ...      ...

As you can see from the table above, daily prices of AMZN and AAPL are different but the Interest and Unemployment rates are the same. Can I feed the data to my neural network like the table above?

In other words, can I put the individual stocks' information besides the environment feature such as interest rates?

Answer 1

I am curious to know if I can put the economy feature such as 'national interest rate' or 'unemployment rate' besides each stocks' features.

The variables are macro-econometric and they, in general, seem to have some influence on stocks' prices. This inclusion might as well increase your model's prediction accuracy. You can definitely use them as predictors. As mentioned in comments - Experimentation is the way to go.

Can I feed the data to my neural network like the table above?

In general, you can have any kind of numeric variables as input to a neural network. Things will work out fine. The important thing is the selection of relevant predictor variables that, potentially, have some relationship with the response variable.