FIX: tiny prediciton intervals

[martins0n]

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

sem is used for variation estimation of sample mean, but we're trying to estimate interval for new sampled value so we should standard deviation intead

Related Issue

Closing issues

[github-actions]

🚀 Deployed on https://deploy-preview-722--etna-docs.netlify.app

[codecov-commenter]

Codecov Report

Merging #722 (cabece0) into master (901c87a) will decrease coverage by 0.00%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #722      +/-   ##
==========================================
- Coverage   83.68%   83.67%   -0.01%     
==========================================
  Files         120      120              
  Lines        6630     6629       -1     
==========================================
- Hits         5548     5547       -1     
  Misses       1082     1082              

Impacted Files	Coverage Δ
etna/pipeline/base.py	94.44% <100.00%> (-0.03%)	⬇️

📣 Codecov can now indicate which changes are the most critical in Pull Requests. Learn more

[martins0n]

from etna.datasets import generate_const_df, TSDataset
from etna.models import NaiveModel
from etna.pipeline import Pipeline

from copy import deepcopy
from etna.analysis import plot_forecast

df = generate_const_df(periods=200, start_time="2021-01-01", scale=0, n_segments=2, add_noise=True)

ts = TSDataset.to_dataset(df[["target", "segment", "timestamp"]])
ts = TSDataset(ts, freq="D")
ts, ts_test = ts.train_test_split(test_size=10)

pipe = Pipeline(
    model=NaiveModel(),
    transforms=[
    ],
    horizon=10
)
pipe.fit(deepcopy(ts))
forecast = pipe.forecast(True)

plot_forecast(
    forecast, test_ts=ts_test, train_ts=ts, prediction_intervals=True, n_train_samples=100
)

Before:

After:

[iKintosh]

Thanks