This is an old revision of the document!
Table
Tables present simulation output data in tabular form. Data can be organized in different ways depending on the dimension of the data.
For instance, a table might display concentrations of three contaminants, for ten time-points. This data can be arranged by time (rows) and contaminants (columns):
| Time | Nickel | Lead | Silver |
|---|---|---|---|
| 0.0 | 1 | 2 | 3 |
| 5.0 | 2 | 8 | 5 |
| … | … | … | … |
| 100.0 | 3 | 9 | 5 |
by contaminants (rows) and contaminants (columns):
| Time | 0.0 | 5.0 | … | 100.0 |
|---|---|---|---|---|
| Nickel | 1 | 2 | … | 3 |
| Lead | 2 | 8 | … | 9 |
| Silver | 3 | 5 | … | 5 |
If the data is probabilistic the dimension increase to three. With n=1000 iterations, the table would then become:
| Time | Nickel/1 | Lead/1 | Silver/1 | Nickel/2 | Lead/2 | Silver/2 | Nickel/n, | Lead/n | Silver/n | |
|---|---|---|---|---|---|---|---|---|---|---|
| 0.0 | 1 | 2 | 3 | 4 | 3 | 2 | … | 5 | 3 | 2 |
| 5.0 | 2 | 8 | 5 | 2 | 6 | 1 | … | 8 | 6 | 8 |
| … | … | … | … | … | … | … | … | … | … | … |
| 100.0 | 3 | 9 | 5 | 6 | 2 | 4 | … | 9 | 6 | 5 |
Reduction
For probabilistic data, it is more common to display statistics rather than data for each iteration. For example, by showing the mean and standard deviation the probabilistic dimension is reduced from thousands of items to just two:
| Time | Nickel/mean | Lead/mean | Silver/mean | Nickel/std dev | Lead/std dev | Silver/std dev |
|---|---|---|---|---|---|---|
| 0.0 | 3 | 1 | 3 | 7 | 9 | 4 |
| 5.0 | 7 | 3 | 1 | 2 | 6 | 3 |
| … | … | … | … | … | … | … |
| 100.0 | 5 | 3 | 6 | 1 | 8 | 5 |
The same can be done for time: instead of showing all time points, just show the first and last time point:
| Time | Nickel/mean | Lead/mean | Silver/mean | Nickel/std dev | Lead/std dev | Silver/std dev |
|---|---|---|---|---|---|---|
| First | 3 | 1 | 3 | 7 | 9 | 4 |
| Last | 5 | 3 | 6 | 1 | 8 | 5 |
If only the last time point is shown, the dimension of the data is reduced so that we can use rows for contaminants instead:
| Contaminant | mean, t=100 | std dev, t=100 |
|---|---|---|
| Nickel | 5 | 1 |
| Lead | 3 | 8 |
| Silver | 6 | 5 |
Depending on which view of the output data you seek there are different tables to choose from:
- Time table - Displays data for each time point. Useful for time dependent simulation outputs, like those from compartments.
- Index table - Displays data for each index. Useful for one or two dimensional outputs.
- Compare table - Displays the ratio, difference or absolute difference between two outputs. For probabilistic simulations or sensitivity analysis only
- Statistics table - Displays data with different statistics.
- Correlation table - Displays the correlation coefficients for one or more outputs.
- Raw data table - Displays all data from a probabilistic simulation for a given time point.
See also
