Case study

1. Dataset

The format for the input sample data file is very straightforward. The minimum contents of it are two variables: date and the value of sample. The daily PM_2.5 observed at US Embassy in the 423-day period from 10 May 2010 to 6 December 2011 is chosen as the sample data in this case. The text input file format is illustrated in Fig. 1-1, which shows the partial contents of the PM_2.5 sample data in the csv file PM25.csv. The reference data PM10.csv contains PM₁₀ observation in 18 official stations, published daily by the Beijing Municipal Environmental Protection Bureau (BJ-EPB) in the same time period with PM_2.5 observation shared in Twitter. Fig. 1-2 shows some of the reference data.

Fig. 1-1 Table of sample data

Fig. 1-2 Table of reference data

2. Exercise

Step 1: The input data format should be similar to the above corresponding data format. Click to import sample data and reference data respectively (see Fig. 2-1).

Fig. 2-1 SPA Model

Step 2: After loading the required data, as in Fig. 2-2, click “Run” button to execute the SPA model and save the result in CSV format. The computed result includes 6 variables: date, area estimation, model weight parameter w₀, the standard deviation STDEV, one standard variance interval CI1 and CI2, shown in Fig. 2-3.

Fig. 2-2 Load data

Fig. 2-3 Computed area estimation and model parameters