Corona Prophet Prediction

[UPDATES] https://www.nature.com/articles/d41586-020-00154-w?utm_source=facebook&utm_medium=social&utm_content=organic&utm_campaign=NGMT_USG_JC01_GL_Nature&fbclid=IwAR3I1vxjD05wwXbGYzqt9jnXVPE6pUiQzbTjISHT6W-niFs5MistDkL2l80

2019-nCoV Global Cases by Johns Hopkins

https://gisanddata.maps.arcgis.com/apps/opsdashboard/index.html#/bda7594740fd40299423467b48e9ecf6

Can we build an original, comprehensive solution to help handle the crisis?

We are looking to predict, visualize and act to contain the 2019 n-Coron Virus, through the power of data science.

With the objective of understanding and minimizing epidemic spread, we devloped an method to accuractely predict and visualize the necessary features relating to n-Corona virus using Time Series Analysis and EDA.

The following dataset has been taken from Novel Corona Virus 2019 Dataset: along with custom feature engineering by extracting data from web.

Kernels

• Mathematical Simulation of nCOV Transmission Model.
• Feature engineeing and analysis of their effects in predicting extent of the nCOV Virus.
• Prediction model for expected new cases in the Mainland China region.
• Prediction model for expected new cases in the Mainland China region.

Mainland China* Death Trends

Following trend indicates daily,weekly,monthly trend in confirmed death rates in China .It was obtained using Prophet an open source tool for Time Series analysis.

*Mainland China includes SAR provinces and Hong Kong.

Uncertainty in Prediction

Prediction with uncertainty and changepoint

Long-term forecast and trends

Correlation Matrix

Pearson’s correlation coefficient is the test statistics that measures the statistical relationship, or association, between two continuous variables. It is known as the best method of measuring the association between variables of interest because it is based on the method of covariance. It gives information about the magnitude of the association, or correlation, as well as the direction of the relationship.

Features include StockPrice,Lowest/Highest Daily Temperature,Humidity(%),StockPrice,Currency,search terms such as cold,etc.

Feature Relationships

We explore the relationship of various features with the spread of the disease by plotting graphs.

• Stock Price Relationship

• Humidity Relationship

• Flights Search Relation

• Lowest Temperature Relation

Simulations

References and citations

[1] https://www.kaggle.com/sudalairajkumar/novel-corona-virus-2019-dataset

[2] https://www.cdc.gov/coronavirus/2019-ncov/index.html

[3]https://gisanddata.maps.arcgis.com/apps/opsdashboard/index.html#/bda7594740fd40299423467b48e9ecf6

[4] https://www.who.int/emergencies/diseases/novel-coronavirus-2019/technical-guidance

[5]https://www.maa.org/press/periodicals/loci/joma/the-sir-model-for-spread-of-disease-the-differential-equation-model

[6]https://facebook.github.io/prophet/

[7]https://ai.googleblog.com/2017/07/facets-open-source-visualization-tool.html

[8]https://scikit-learn.org/stable/

[9]https://seaborn.pydata.org/

[10]https://colab.research.google.com/

[11]https://github.com/wuhan2020/wuhan2020