README- Predicting-Energy-Values

BRIEF DESCRIPTION:

• Here we are basically doing Time Series Forecasting of May month by using ARIMA Model.
• Here I used a modified version of the Dataset called ‘final.csv’

---

PREREQUISITES:

• written in Python 3.6 .

---

CLIENT-END FULFILMENTS:

Steps to predict:

1. Run- ‘1_final_data-loading.py’ file.
   • By running it actually we will consider April & May values for modelling our problem & May values for validating our prediction.
   • 2 files will be created final_dataset.csv & final_validation.csv
2. Run- ‘2_final_problem-evaluate.py’
   • We will evaluate the problem by Baseline prediction & by Drawing various Plots.
   • Output given in jpeg.
3. Run- ‘3_final_arima-modelling.py’
   • Now we will implement ARIMA Model on the Dataset.
   • Here the program will automatically find the best (p,d,q) values for the Problem.
   • Output given in jpeg.
4. Run- ‘4_final_arima-residualerrors.py’
   • Here checking for Residual Errors to see whether Stationarity in Dataset is acheived or not.
   • Output given in jpeg.
5. Run- ‘5_final_model-save.py’
   • Now we will implement the model on entire dataset & save it as ‘final_model.pkl’ & to use it for future predictions.
   • Output given in jpeg.
6. Run- ‘6_final_predict.py’
   • Now this will finally predict the values of the May month along with what was Expected & what it predicted.
   • Output given in jpeg.

---

OUTPUT SAMPLE:

• Arima Modelling Output

• Arima Residual Errors Output

• Final Predictions

---

AUTHORS:

• coded by AAYUSH GADIA.

---

• Click here to visit the GitHub Repo