All Projects
Time Series

Household Power Consumption Forecasting

Multi-model time series on 2.9M UCI records (2006–2010). ARIMA, SARIMA, Prophet, LSTM on Global_active_power. STL reveals daily+weekly patterns. Ensemble with inverse-RMSE weighting across all models.

Dataset

UCI Household Power: 2.9M measurements, 1-min resolution, 4 years

Approach

STL decomposition → stationarity → ARIMA/SARIMA/Prophet/LSTM → ensemble

Tech Stack
PythonTensorFlow/LSTMstatsmodelsProphetPandas
Keywords
LSTMARIMASARIMAProphetSTL DecompositionPower ForecastingUCI
Visualizations6 Charts
Deep Dive

Time series forecasting for household electricity on the UCI dataset.

Dataset

  • 2,075,259 measurements (Dec 2006 → Nov 2010), 1-minute resolution → resampled hourly
  • 1.25% missing values → linear interpolation
  • Target: Global_active_power (kW)

Time Series Analysis

  • ADF test: stationary after differencing
  • STL decomposition: daily pattern (morning/evening peaks) + weekly cycle (weekend dips)
  • ACF/PACF: clear lag-24h and lag-168h autocorrelations

Models Evaluated

ModelStrengths
ARIMAAuto p,d,q selection, AIC
SARIMASeasonal period=24h
ProphetTrend + yearly + weekly + daily seasonality
LSTM (encoder-decoder)24-step ahead sequential prediction
EnsembleInverse-RMSE weighted combination

Key Insight The seasonal component explains ~60% of variance. People follow routines — a "same time yesterday" baseline is hard to beat. Advanced models target the irregular residual component (vacations, guests, appliance failures).

STL Decomposition Findings Trend: slowly declining over 4 years (efficiency improvements). Seasonal: strong daily (double peak: 7–9am + 6–9pm) + weekly (weekends 15% lower). Residual: 1.25% anomalies correlate with holidays.