Linear regression using Matlab

Matlab recognises the importance of linear regression calculations and introduced a very elegant and useful notation the / forward- and backslash operators, see p.l 17-118. 

We have encountered Excel s LINEST as a tool for linear regression. Unfortunately, LINEST cannot be generalised from vectors to matrices. To deal with matrices, we do not have an option but to use equations (4.59) and (4.61). It is possible to do so, but not as convenient as in Matlab. 

Other software packages that can be used to perform non-linear regression analysis are MathCad (MathSoft, Inc., 201 Broadway, Cambridge, MA 02139) and MatLab (Mathworks Inc., 24 Prime Park Way, Natick, MA 01760). 

From experimental data (file expdata.dat available on the worldwide web), we want to determine the best value of the product of the overall heat transfer coefficient times the heat transfer area (UA). You need to use the load command which will give you a two-dimensional array with time and temperature called exp-data. Is there a linear plot that allows for the use of linear regression MATLAB has the function polyfit that can perform this fit. What is the optimal vaue of UA for this system What do you think of the data MATLAB also has a function polyval that allows you to get best predicted values. 

SVM Toolbox, http //asi.insa-rouen.fr/%7Earakotom/toolbox/index. This fairly complex MATLAB toolbox contains many algorithms, including classification using linear and quadratic penalization, multiclass classification. E-regression, v-regression, wavelet kernel, and SVM feature selection. 

---