Time Series Analysis - Statistics Solutions

Time Series Analysis

<aside> 💡 Check the stationarity of the data. A dataset is called stationary if the mean and variance is same for a given period. If the data is not stationary then make it stationary by appropriate process. This might be a reason for the the poor performance of our data in traditional method.

</aside>

Paper Organization

Introduction

Time Series Analysis

Demand Forecasting in Smart Grid

Data Analysis

Comparison among different models

Conclusion

https://www.youtube.com/watch?v=Prpu_U5tKkE

https://www.youtube.com/watch?v=DeORzP0go5I

Understanding LSTM and its diagrams

tso mae = 320.67740705418134
Test Mean Absolute Error: 3729.264425078833 (AR)
Test Mean Absolute Error: 3862.929129434021 (ARMA)
Test Mean Absolute Error: 3865.2968494627607 (ARIMA)
Test Mean Absolute Error: 361.40645029085636 (LSTM)