During Monte Carlo simulation data generated is a lot and non-trending. Is there any member to remember previous ways to sort out relevant data and keep outliers away as well as any tool to furnish the results' presentation?
The question is so broad as to make answering difficult. As you know, Monte Carlo techniques are typically used in combination with a simulation to produce artificial data. As such if you do not get data similar to what you see in your instruments here are a few simple things to check:
1. Are you simulating what you think you're simulating?
Some codes have the ability to simulate a wide variety of processes, but you may wish to only consider a single process. Are the other processes turned off?
2. For the processes you are simulating, are the settings appropriate?
Most simulation code, in order to make efficient use of computer resources apply cut-offs in critical parameters such as energy below or above which the data are uninteresting for your purpose. If you've chosen cut-off values to high or low, you will likely not get the answer you expect.
3. Are you simulating the instrument used to detect the process of interest correctly? Is the geometry correct? Are the materials correct? Is the surrounding environment correct? Do you include instrumentation effects such as threshold, gain, response function, etc.?
4. If you have real data (or have experience with real data) do you apply any selection to those data? If so, you want to apply the same selection to your artificial data.
Look at the site http: //root.cern.ch. Probably the toolkit described there will be useful to you.
- Badges
- Science topic
- Similar topics
- Engineering
- Modeling