The aim of the paper is to point out that the Monte Carlo simulation is an easy and flexible approach when it comes to forecasting risk of an asset portfolio. The case study presented in the paper illustrates the problem of forecasting risk arising from a portfolio of receivables denominated in different foreign currencies. Such a problem seems to be close to the real issue for enterprises offering products or services on several foreign markets. The changes in exchange rates are usually not normally distributed and, moreover, they are always interdependent. As shown in the paper, the Monte Carlo simulation allows for forecasting market risk under such circumstances.
The purpose of this study is to identify relationships between the values of the fluidity obtained by computer simulation and by an experimental test in the horizontal three-channel mould designed in accordance with the Measurement Systems Analysis. Al-Si alloy was a model material. The factors affecting the fluidity varied in following ranges: Si content 5 wt.% – 12 wt.%, Fe content 0.15 wt.% – 0.3wt. %, the pouring temperature 605°C-830°C, and the pouring speed 100 g · s–1 – 400 g · s–1. The software NovaFlow&Solid was used for simulations. The statistically significant difference between the value of fluidity calculated by the equation and obtained by experiment was not found. This design simplifies the calculation of the capability of the measurement process of the fluidity with full replacement of experiments by calculation, using regression equation.
When an artificial neural network is used to determine the value of a physical quantity its result is usually presented without an uncertainty. This is due to the difficulty in determining the uncertainties related to the neural model. However, the result of a measurement can be considered valid only with its respective measurement uncertainty. Therefore, this article proposes a method of obtaining reliable results by measuring systems that use artificial neural networks. For this, it considers the Monte Carlo Method (MCM) for propagation of uncertainty distributions during the training and use of the artificial neural networks.