tunino91 / stockPredictionRegression

Linear regression classifier for GOOGL ticker

Stock Prediction with Linear Regression

Used Features

1. Adj. Close: Close price of the day
2. HL_PCT: High to low percentage change (High - Low) / Low
3. PCT_change: Daily percentage change (Close - Open) / Open
4. Adj. Volume: Daily traded volume

Cleaning

• All the NaNs are filled with -9999 to count as an outlier by the sklearn LinearRegression object's fit function.

Adjustable Parameter

• Ticker is currently set for GOOGL, you can pick another as you please.
• You can determine how much of historical data you want to incorparate in your predictions by changing the forecast_out variable