The completeness of data (both in terms of how often it is updated as well as its representativeness) that Prospection receives varies depending on the region and data source.
To provide a more recent, representative and usable view of the current market Prospection applies analytic solutions to the data including:
- Scaling: extending the data to represent the broader population (recency and/or representativeness), or;
- Modeling: analysing historical data to predict incomplete data (recency)
Depending on the dataset, using the Modeled feature will apply one or both of these in order to make the data more complete.
What does this feature show you?
Using the toggle above the chart, you can select to display actual data or modeled data in the chart and data table within the report.
Actual data: Displays the raw data with no analytic corrections applied to improve data recency and/or representativeness
Modeled data: Displays the predicted data by applying scaling and/or modeling analytic solutions to provide more recent and/or representative data
Predictive Intervals
When you hover over a data point that is a ‘predictive’ value (modeled data), predictive intervals are included. Displayed are the lower and upper bound of the estimates for the predictive value we have provided. If there is not enough historical data to calculate an estimate, this will be displayed.
Note: Predictive intervals are only applicable when the modeling solution has been applied to the data.
Exporting to CSV. file
When exporting data tables to a CSV file, the display option selected (actual data or modeled data) will be displayed in the tables. If there is not enough historical data to calculate an estimate for modeled data this will also be displayed.
Relevant reports
The actual and modeled data feature is available in the following reports:
- Patients on Treatment
- Patient Dynamics
- Medication Counts