Sergio’s research presents, for the first time, the time to independence data technique applied to continuous emission data. The results obtained from its use on NOx and CO2 emission data show that this technique is most useful and effective in mitigating autocorrelation. This technique is relevant given the advancements in sampling devices and data collection capabilities that have afforded the collection of enormous amounts of data for a myriad of purposes. These capabilities include smart meters, digital pedometers, medical devices, and continuous emission monitoring systems in factory stacks among many more. We can currently measure every instant of virtually every activity thanks to the advances in technology. That is why data handling techniques like the time to independence are most necessary in finding meaning out of the colossal amounts of data at our disposal.
The time to independence method described is a valuable tool that can make any subsequent statistical analysis valid and robust since autocorrelation in the data would be mitigated. Thus, once data are composed of quasi-independent observations, a more meaningful statistical analysis may ensue since the correct use of an ANOVA or GLM analysis will be warranted. Under such an analysis the significance of independent variables can be determined, allowing then for the testing of the significance such as fuel types, engine parameters, and ambient parameters on engine emissions.
To learn more feel free to read more in Sergio’s dissertation: Novel Data Analysis Technique to Evaluate Field NOx and CO2 Continuous Emission Data, Based on the Evaluation of: (1) An Off-Road Diesel Compactor Running on three Fuel Types and (2) Two Compactors Running on Diesel Fuel