Skip to content

Benchmarking of different algorithms used fot time series forecasting ARIMA, Prophet, LSTM

Notifications You must be signed in to change notification settings

amine-akrout/Time-series-forecasting

Repository files navigation

Time series Analysis

Data used in the Notebooks can be found here : https://archive.ics.uci.edu/ml/datasets/Metro+Interstate+Traffic+Volume

Data Set Information:

Hourly Interstate 94 Westbound traffic volume for MN DoT ATR station 301, roughly midway between Minneapolis and St Paul, MN. Hourly weather features and holidays included for impacts on traffic volume.

Attribute Information:

  • holiday Categorical US National holidays plus regional holiday, Minnesota State Fair
  • temp Numeric Average temp in kelvin
  • rain_1h Numeric Amount in mm of rain that occurred in the hour
  • snow_1h Numeric Amount in mm of snow that occurred in the hour
  • clouds_all Numeric Percentage of cloud cover
  • weather_main Categorical Short textual description of the current weather
  • weather_description Categorical Longer textual description of the current weather
  • date_time DateTime Hour of the data collected in local CST time
  • traffic_volume Numeric Hourly I-94 ATR 301 reported westbound traffic volume

Explorary Data Analysis :

univariate analysis, seasonality and correlations

Benchmarking of different Algorithms used for forecasting :

Algorithm Training Time Forecasting Time RMSE MAE MAPE
ARIMA 638 ms 1min 11s 542.68 394.21 23.61 %
Prophet (baseline) 2.6 s 2.01 s 985.96 755.68 40.36 %
Prophet (1 regressor) 1.87 s 2.2 s 842.02 636.90 37.79 %
Prophet (4 regressors) 1.09 s 2.16 s 837.53 626.79 36.02 %
Prophet (optimized) 31.4 s 2.57 s 380.87 293.86 18.40 %
LSTM 1min 31s 628 ms 613.44 426.53 14.62 %
XGBoost (only time features) 1.81 s 9 ms 311.21 213.37 12.48 %
XGBoost (all features) 3.02 s 6.99 ms 281.19 181.95 9.68 %
XGBoost (fine tuned) 5.64 s 5 ms 277.48 181.79 8.00 %

About

Benchmarking of different algorithms used fot time series forecasting ARIMA, Prophet, LSTM

Topics

Resources

Stars

Watchers

Forks