Starting at $500.00

time series forecast and anomaly detection 30 models arima prophet, lstm, etc

jidiv
jidiv
0 order in queue

5 days for delivery

N/A
( 0 reviews )
Contact seller

Actions

Complete analysis and prediction of Time series data, using Python. Includes:

  • Google Colab Notebook (web app, no software to install) containing the entire python pipeline that reads data, generates predictions, shows interactive charts, exports output
  • Documentation of the theory behind each step included in the notebook
  • Replicability/Scalability: Entirely reusable with other data
  • Methodologies: Up to 30+ different Time Series & Machine Learning algorithms


The delivery contains:

  1. Time Series Decomposition into trend, Seasonal (yearly/weekly/daily/holidays) & Error components
  2. Anomaly Detection: Identify anomalies (and their importance) in the past
  3. Statistical tests to check stationarity, normality, etc + ACF/PACF charts
  4. Fit TS Models: Create 30+ TS models (Arima, Prophet, ETS, GARCH,,...) compare performance, select the best performing one
  5. Hyperparameter optimization of best TS Model
  6. Forecast: Compute Future Forecasts
  7. Type of modeling:
  • Univariate: use past values of target variable (y)
  • Multivariate: Use past values of target + exogenous variables (x1, x2, etc.)
  • Single or Multiple TS: MAke the forecast for a single time series, or for multiple TS in parallel (e.g. multiple SKUs)


You may also like

You may also like the following gigs