AQI-Prediction-before-and-after-Lockdown-of-COVID-19-in-India using SARIMA and LSTM

1. The study analyses the AQI and predicts before and after lockdown of COVID-19 in India. The values for the future is also predicted using the SARIMA model and LSTM algorithm.

2. SARIMA stands for Seasonal Autoregressive Integrated Moving Average model is capable of analysing and representing stationary and non-stationary time series by using optimal hyperparameters for better prediction.

Fig 1. Seasonal decompose of AQI values of Delhi.

Fig 2. Analysis of the Pollutants in various years

Fig 3. Comparison of AQI before and after lockdown

Fig 4. Test and Train prediction data of AQI for Delhi

Fig 5. Forecasted AQI values of Delhi for the year 2020–2021 (a) SARIMA (b) LSTM

3. The performance evaluation of the forecasting models by calculating mean square error, mean absolute error, and root mean square error (RMSE) which may help control the degraded air quality.