Q:  Why does data analysis need to be done?
A:  If it is worth your time to collect the data, then it is worthwhile to see what the data reveals (data revelation!)

Q:  What is statistical modeling, and what is it good for?
A:  It creates an algebraic equation that fits the data. This equation can then be used to predict the future or optimize the system that created the data.  It will tell the user what the value of (Y) will be if one or more of the independent variables (X1, X2, X3, etc.) is changed.

Q:  Is the analysis sophisticated enough for professional statisticians?
A:  Absolutely.  We believe DataRevelation
generates a world-class model.  It is unlikely that any competing model, with the same number of terms, will fit the data better.

Q:  Will DataRevelation
work if there is only one independent variable?
A:  Sure.  This two dimensional problem is often called curve fitting, and it can be considered a relatively simple subset of multivariate statistical modeling.. 

Q:  Is DataRevelation
the best software for scientific fields where the model is well known with only one or two variables?
A:  It may show the user something new, and it is probably easier to use than other software.  We believe DataRevelation is outstanding with more mathematically complex problems where there are many independent variables, it is not known which are important, and the mathematical relationships between the dependent and independent variables are not known.


Q:  What is the coefficient of determination (R-squared) and what is it good for?
A:  It is a number between 0 and 1 that indicates how good the model (equation) fits the data.  An R-squared of 1 is a perfect fit, and the dependent variable (Y) will be predicted perfectly for every value of the independent variables.

Q:  What is a dependent and independent variable?
A:  The dependent variable is the (Y), and it is the value you want to predict.  The independent variables (X1, X2, X3, etc.) are the things that can be varied in your system.

Q:  Why is caution needed when using two term polynomials (X + X**2)?
A:  It describes a parabolic function that can be manipulated to fit many curves.  The danger is the parabolic functions have drastic motion outside of the data.  If the user wants to use the model near the edges of the data (or outside the data), then do not use two term polynomials.  Models are usually more accurate near the mid-point values of the independent variables because this is where the data is.

Return to Home Page