BShade Prediction

Feature Description

BShade Prediction is the statistical inference process using the BShade spatial sampling model for biased samples.

Feature Entry

  • Spatial Statistics Tab -> Spatial Sampling and Statistical Inference -> BShade Prediction。
  • Toolbox->Spatial Statistics->Spatial Sampling and Statistical Inference-> BShade Prediction。

Parameter Description

Source Dataset Field: Specify the data field in the source dataset. In the case below, this shows 5 selected fields recording data from sample hospitals.

Historical Data Field: Specify the data field in the historical dataset. In the case below, this shows 19 selected fields recording last year's hospital data.

  • Source Dataset: Set the dataset and its datasource for BShade statistical inference. In the case below, this represents this year's morbidity data from 5 sample hospitals.
  • Historical Data: Specify the historical dataset and its datasource. In the case below, this represents last year's morbidity data from 19 hospitals.
  • Parameters:
  • Estimation Method: BShade estimation method. Total method (based on sample-to-population ratio) and Average method (based on sample mean-to-population mean ratio).
  • Result Data: Set the result dataset and its datasource.
  • Click "Execute" to run the analysis. After completion, the output window will indicate whether the result succeeded or failed.

Application Case

A region needs to predict daily incidence of a disease this year. With historical data from designated hospitals and daily incidence data from 5 BShade-sampled hospitals, BShade Prediction can forecast the total daily cases during this period.

  • Case Data: Click here to download BShade sampling and prediction case data. Hospital_Case_Historical contains historical incidence data from all regional hospitals; Sentinel_Case contains daily incidence data from 5 sample hospitals.
  • Parameters: After downloading the data, open BShadeData.udbx and configure parameters as shown below to execute BShade prediction analysis.
  • Result Interpretation: The predicted daily disease incidence results for the region are shown below: