Air Quality Archives – Neil R. Gimson

Air Quality Predictions from Weather Data – Part 3

Logistic Regression Model: The simple models developed in Part 2 to classify airborne fine-particle concentration in terms of local meteorology are improved on using logistic regression techniques.

Air Quality Predictions from Weather Data – Part 2

Cluster Analysis of Meteorological Features: The exploratory data analysis of Part 1 is continued, to include a cluster analysis and develop simple models which classify the fine-particle concentration according to the local meteorology.

Air Quality Predictions from Weather Data – Part 1

Exploratory Data Analysis: This set of posts develops statistical models to predict levels of fine particles in Christchurch’s air, using weather parameters such as wind, temperature and humidity.

Simulations of Future Air Quality – Summary

This summary outlines a model of air-pollution concentration in central Sydney. Technical details of the model development are contained in previous posts on this subject.

Simulations of Future Air Quality – Part 3

Future daily pollution concentrations are simulated in central Sydney, using a periodic STL decomposition of recent time-series data and an ARIMA model for the seasonally-adjusted component.

Simulations of Future Air Quality – Part 2

Several forecasting models, including ETS and ARIMA, are applied to monthly averages of air-pollution data in Sydney. The models provide forecasts of the averaged seasonal component but are not suitable for simulations, as the seasonal pattern erodes away with time. Further refinements to better simulate the daily average of the seasonal data are described in a subsequent post.

Simulations of Future Air Quality – Part 1

Air-pollution concentrations at a monitoring site in central Sydney are projected into the future using a simple block bootstrap method, which samples recent site data. The simulated time series follows the current distribution of concentrations; refinements to capture the auto-regressive properties of the data are described in subsequent posts.