This is the code used in the second application of the paper Residual-based CUSUM beta regression control chart for monitoring double-bounded processes from Cristine Rauber, Luiz M. A. Lima-Filho, and Fábio M. Bayer.
The dataset refers to the relative humidity (RH) in Australia and highlights the relevance of the proposed chart in monitoring double bounded environmental data. The RH is a ratio between continuous numbers, being the ratio of the partial pressure of water to the equilibrium vapor pressure of water, assuming values in (0,1). Due to the genesis of the beta regression model, rates and proportions usually can be well fitted by this model. Additionally, the monitoring of RH is relevant because it exerts influence on temperature, rain, and thermal sensation. This dataset is available in the rattle package in R. However, the data has been updated so the results presented here do not reflect the results in the paper.
| Feature | Description |
|---|---|
| RelHumid3pm | Relative humidity (%) at 3pm |
| Cloud3pm | Fraction of sky obscured by cloud at 3pm |
| Evaporation | The so-called Class A pan evaporation (mm) in the 24h to 9am |
| MaxTemp | The maximum temperature in degrees celsius |
| MinTemp | The minimum temperature in degrees celsius |
| Pressure3pm | Atmospheric pressure reduced to mean sea level at 3pm |
| Rainfall | The amount of rainfall recorded for the day in mm |
| Sunshine | The number of hours of bright sunshine in the day |
Rauber, C., Lima‐Filho, L. M., & Bayer, F. M. (2022). Residual‐based CUSUM beta regression control chart for monitoring double‐bounded processes. Quality and Reliability Engineering International.
Cristine Rauber (rauberoliveira at gmail.com)