Implement other code examples on Predictive Models which detect anomalies using MAPE/MSE value

[JoeRishon]

We have already implemented one code sample which uses fbprophet to detect anomalies We would want you to make more code examples using other predictive models. :pray:

List of Predictive Models that can be used

• [x] ARIMA

• [x] SARIMA

• [x] ARCH

• [ ] GARCH

• [ ] VAR

• [ ] VECM

• [ ] LSTM

• [ ] Convolutional Autoencoder model

Instructions to be followed

• The Code examples on the implementations of most of the models mentioned above should be available in the internet. :octocat:
• Build the respective predictive model on the NAB Dataset(RealAWSCloudwatch - Any CPU Utilization Data ONLY).
• Use the Predictive Model to forecast future values.
• Find the Error rate of the predictive (Use MAPE/MSE) and come up with a confidence interval for the predicted value.
• Any Actual datapoint falling beyond the confidence interval, will be detected as an anomaly.

Additional instructions

• Make a new jupyter notebook with the algorithm name and work in that only.
• Initially above algorithms will be assigned on a First Come First Serve basis. The algorithm will be assigned to you if you create an "assign request" comment below with a small description of the algorithm you have requested for. :smiley: :1st_place_medal:

@JoeRishon & @manikyabard for any doubts and additional details

[RakshithRAcharya]

Hey @manikyabard and @JoeRishon can I do this??