Random numbers, variance reduction & Software

10 important questions on Random numbers, variance reduction & Software

Which methods are explained to derive random numbers from any continuous distribution?

1. Inverse transformation method: Calculate F-1(U)
2. Convolution method.

What is the aim of variance reduction?

Variance reduction aims to reduce the uncertainty in the simulation output by deliberately introducing useful correlation in the simulation output.

What are the (dis)advantages of variance reduction?

-Speeds up experimentation; less runtime needed
- But extra programming effort is required
  • Higher grades + faster learning
  • Never study anything twice
  • 100% sure, 100% understanding
Discover Study Smart

Why do common random numbers decrease the variance over the differences of two systems?

The use of CRN induces correlation between the experiments. This leads to a co-variance term that reduces the total variance.

What is the meaning of synchronization within the use of CRN?

Random numbers across different system configurations are matched. (Each activity has the same R.N regardless of the configuration)

Which methodology is applied with the Antithetic Variates Technique?

Use negative correlation to reduce the variance of one single configuration

What is the idea behind simulation based optimization?


Sequentially decide on the system configuration to simulate (value of the experimental factors) based on the results from earlier simulated configurations.

Which 5 requirements for simulation software are mentioned in the sheets?


1.Modeling capabilities
2.Animation
3.Statistical capabilities
4.Output reporting & plots
5.Support & maintenance

What are the (dis)advantages of graphical software?


Pros
-Better understanding of the decision maker of the decision options, the system to be modelled and the model itself.
-Assists the modeller in understanding the system / problem and in debugging.
-Better communication decision maker Ûmodeller.
-Simpler model validation.
-Provides different and better insights.
Cons
-Danger of putting too many details in the model (model too complicated, which costs too much development and debugging time, without providing a lot of additional insight)
-Danger to put too much emphasis on the outer appearance (animation) rather than on the model structure (example: 3D animation)

What is a disadvantage of the empirical test on random number generators?

It only provides an insight in the segment of the cycle that was actually tested.

The question on the page originate from the summary of the following study material:

  • A unique study and practice tool
  • Never study anything twice again
  • Get the grades you hope for
  • 100% sure, 100% understanding
Remember faster, study better. Scientifically proven.
Trustpilot Logo