Custom «Computer Programs» Essay Paper Sample

Computer programs use an algorithm to generate random numbers. That means that it is possible for anybody to predict what numbers will result if he or she knows the algorithm. These numbers are pseudorandom numbers because they are not truly random. Therefore, pseudorandom numbers are useful in simulations because it is possible to predict the resultant numbers. Pseudorandom numbers are useful in a Monte Carlo method simulation that refers to as a method that makes use of sequences of random numbers to do the simulation. The calculation of the Pi (π) is an example of how a computer performs a Monte Carlo simulation.

Pseudorandom numbers and the accuracy of a simulation

The pseudorandom numbers are extremely accurate because of the tendency of predicting the results. For instance, the Monte Carlo simulations are applicable in various areas such as cancer radiation therapy, stellar evolution, traffic flow, and quantum chromo dynamics, which indeed depend on high accuracy. Pseudorandom numbers ensure that these simulations are extremely accurate.

Role of statistical analysis in of simulation

Statistical analysis plays a significant role in simulation, including predicting numbers that will occur in the future.

Exponential smoothing forecast method

The exponential smoothing forecast method is among the significant forecasting methods that are useful in the real life situation. This forecasting method weighs the most recent past data more strongly than it does to the distant past data. Therefore, the forecast will be more effective to immediate changes in the data, which is good to study when dealing with seasonal trends and patterns that may be taking place. This information will be significant when studying the increased production of a commodity that seems to be in higher demand during the recent times than in the past.

Using a forecasting model

The domestic fast food chain would consider the data of other companies  put together a forecast to determine the viability of their business venture. The domestic fast food chain could examine the sales data and use the exponential smooth forecast to determine if it is sensible for them to enter into the market. However, this would only show the changes in the data that are more recent than data in the past time.

The difference between a casual model and a time-series model

The time-series model makes use of the historical data in determining the future behavior. This model could be useful to fast food restaurants, clothing manufacturers, or retail stores to determine sales for a forthcoming season change. On the other hand, the casual model depends on the use of a mathematical correlation between the predicted items and the factors that affect how the predicted items behave. The casual model would be useful for those companies that use the competitors’ available data because they do not access the historical data.    

Problems of the moving average forecasting model

The most common problem with the moving average forecasting model is that it does not consider the data that change because of seasonal trends and variations. This model is useful only in the short run forecast but is not reliable in forecasting events two far in the future.

Determining the number of observations to average in a moving average model

Trial and error experimentation is useful when determining the number of observations in the moving average model.

Determining the weightings to use in a weighted moving average model

Trial and error experimentation makes it possible to determine the weightings to apply in a weighted moving average model. If the weightings are not accurate, the outcomes of the model will not effectively predict the future. Therefore, accuracy of the weightings is particularly significant in a weighted moving average model.