Energy Price Forecasting

• 

  Log return graph of prices

What it does

Energy price.ipynb: Cleans the three price datasets, ensuring they line up, imputing missing values, removing erroneous data, merges the three datasets together, adds daily average temperature data Scales and picks features to feed into Machine Learning Algorithm for predicting the impending System Price AI_Hackathon.ipynb: Uses the three price dataset to produce graphs that show the characteristics of each dataset, how they relate to each other, their volatility, seasonality, etc

How we built it

Using pandas, python on GoogleCollab and Jupyter. The histogram image relied on Excel

Challenges we ran into

Dealing with the minute details of the data preprocessing stage took a lot of time

Accomplishments that we're proud of

The data is finally ready to be put into forecast, and the graphs offer great insight into the data

What we learned

Data preprocessing is hard and takes a lot longer than anticipated

What's next for Energy Price Forecasting

Acting on the recommendations listed in the end of the jupyter notebook to improve the forecasting models performance

Built With

• pandas
• python
• scikit-learn