# EuroOffice Modeller - Overview

EuroOffice Modeller turns your spreadsheet into a powerful statistical modelling tool. It contains an extended set of mathematical and statistical functions for solving optimization, data analysis and time series forecasting problems. Installed as an extension of EuroOffice (or any other OpenOffice derivative, such as LibreOffice or Apache Open Office), it integrates seamlessly with existing spreadsheet functionality.

## Optimization

Decision making is often based on finding an optimum among many possibilities.

Solvers, or optimizers, are software tools that help users **determine the best way to do** something. This may involve many topics in economics, education, sciences and everyday life –, for example allocating money to investments, locating new warehouse facilities, scheduling hospital operating rooms, blending foods, diets, planning new products or shipping routes, etc. In each case, multiple decisions need to be made in the best possible way while simultaneously satisfying a number of logical conditions (or constraints). The "best" or optimal solution might mean maximizing profits, minimizing costs or achieving the best possible quality.

With Modeller, you can find an optimal (maximum or minimum) value for a formula in one cell — called the **objective** cell — subject to **constraints**, or limits, on the values of other formula cells on a worksheet. Modeller works with a group of cells — called **decision variable** cells — that participate in computing the formulas in the objective and constraint cells. Modeller adjusts the values in the decision variable cells to satisfy the limits on constraint cells and produce the required (maximal, minimal or a particular, fixed value) result in the "objective cell".

The objective, constraint and decision variable cells and the formulas interrelating them form an Optimization **model**; the values found this way are called a **solution** for this model. We use a variety of methods, from linear programming and nonlinear optimization to stochastic algorithms, to find solutions.

Read more on optimization...

## Statistical Analysis

We are all overwhelmed with data which are meaningless if we cannot find order in them.

Sometimes we have no idea or we just do not want to have any preconception about our data. In such a case the first step is to investigate the general statistical properties (mean, variance, etc. ) of our data set with *Basic Statistics* and have a visual picture of them using a *Histogram*.

Sometimes we have a general idea about what our data may prove. In this case we can run various statistical tests to determine if our assumptions can be justified by the data. *T-Tests* and *F-Tests* can prove if there are differences in the mean and variance of two data sets. Analysis of variance (*ANOVA*) is used to determine the effects of input variables on experiment results.

## Time Series Analysis

Analysing the past may help you to look into the future. Times series analysis methods give you insight in the temporal behaviour of your data. You can separate signal from measurement noise or random errors and you could find repeating patterns that can be extrapolated into the future.

**Regression analysis** relates data to other data, called explanatory variables. If you have a time series of the explanatory variable for a time period where your data is unknown then you can have a good estimate based on the connection between your data and the explanatory variable. The **Kalman-filter** is a well known method for eliminating noise from signal and it can be used everywhere you have measurement data: science, engineering, or even processing survey data of any kind.