Add confidence intervals to SARIMA

[alex-hse-repository]

IMPORTANT: Please do not create a Pull Request without creating an issue first.

Before submitting (must do checklist)

• Did you read the contribution guide?
• Did you update the docs? We use Numpy format for all the methods and classes.
• Did you write any new necessary tests?
• Did you update the CHANGELOG?

Type of Change

• Examples / docs / tutorials / contributors update
• Bug fix (non-breaking change which fixes an issue)
• Improvement (non-breaking change which improves an existing feature)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)

Proposed Changes

Add confidence intervals to SARIMA

Related Issue

Closing issues

[codecov-commenter]

Codecov Report

Merging #172 (1bfbd5d) into master (8d09303) will decrease coverage by 0.81%.
The diff coverage is 97.56%.

@@            Coverage Diff             @@
##           master     #172      +/-   ##
==========================================
- Coverage   87.22%   86.40%   -0.82%     
==========================================
  Files          74       74              
  Lines        3272     3311      +39     
==========================================
+ Hits         2854     2861       +7     
- Misses        418      450      +32     

Impacted Files	Coverage Δ
etna/models/sarimax.py	94.89% <97.56%> (+1.01%)	⬆️
etna/models/nn/deepar.py	75.51% <0.00%> (-24.49%)	⬇️
etna/models/nn/tft.py	76.47% <0.00%> (-23.53%)	⬇️
etna/transforms/pytorch_forecasting.py	92.40% <0.00%> (-7.60%)	⬇️
etna/loggers/base.py	92.30% <0.00%> (-1.93%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 8d09303...1bfbd5d. Read the comment docs.

[iKintosh]

Docs fixes 👍

[iKintosh]

It will work, but may need some changes in the future.

[iKintosh]

Have you considered possibility to use standard forecast method?

As I understand confidence_interval parameter makes it impossible if _forecast_segment is staticmethod, but it is not necessary to keep it static.

[alex-hse-repository]

We can use standard methods in the future when we add ci to the rest of the models, but now Julia advised me to do it this way