What a good software or program to design experiments and find the regression models? Polynomial Models? Linear Regression? design of experiments DOE?
for three or more variables?
1/ curve fitting tools MATLAB
2/ fitlm function from Statistics and Machine Learning Toolbox MATLAB
3/ JMP software
4/ MINITAB software
5/ SPSS software
6/ ..................
If you are purely interested in Design of Experiments, polynomial models and linear regression I recommend Modde: http://umetrics.com/product/modde-pro
Dear Ayoub
SPSS software is the easy and efficiency program because under the window and needs simple training, another program SAS very good but need more training.
Good Luck
Thank you all for your answers
Another question is how much experimental points I need for a good regression model for ten variables?
Dear Ayoub
You can use G*Power software to estimate the required points for multiple regression. From the menu Tests chose multiple regression or the suitable choice for your study and complete the procedure.
G*Power software is an effective tool to calculate sample size for many ranges of experiments. Also, you can determine effect size and power of the test, G*Power is free to download and easy to use after reading the manual, the download link:
http://download.cnet.com/G-Power/3000-2054_4-10647044.html
The link to download G*Power manual:
http://www.gpower.hhu.de/fileadmin/redaktion/Fakultaeten/Mathematisch-Naturwissenschaftliche_Fakultaet/Psychologie/AAP/gpower/GPowerManual.pdf
I hope that be useful for you.
Good Luck
Hi Ayoub,
if you have worked with R, you might find nice tools in the CRAN Task View: Design of Experiments (DoE) & Analysis of Experimental Data. https://cran.r-project.org/web/views/ExperimentalDesign.html
Good luck!
Good topic
Comparison of statistical packages
https://en.wikipedia.org/wiki/Comparison_of_statistical_packages
pl get interesting article in link ;http://www.statgraphics.com/design-of-experiments
STATGRAPHICS contains extensive capabilities for the creation and analysis of statistically designed experiments. Statgraphics functions as design of experiments software that creates designs of several types.
Basically you will find general purpose tools, such as STATGRAPHICS and MINITAB or dedicated DoE software, like DesignExpert or ModdE. Both can work for you. Dedicated tools might be easier to use, but general purpose tools offer a higher flexibility, when it comes to other methods, such as regression. JMP or R are pretty tough to operate, but offer a huge spectrum of methods. Depends on how much you would like to invest. Personally I work most of the time with Statgraphics (since 1999 and still like it).
statistica is the better http://www.statsoft.fr/index.php
This software does really well for me on regression. It can do multiple regression and customizable polynomial. Too bad it doesn't have logistic. But it is pretty new. Maybe they will add that function soon.
You can find it at https://vinci.bioturing.com/
Thank you all
So far I have not found a good software to study 10 factors with 100 experiments
Who has an idea?
any software like SPSS,SAS and even MS Excel may serve your purpose.
Over the past decade, Giustolisi and Savic (2006) have proposed the evolutionary polynomial regression (EPR), which combines the most expedient features taken from traditional numerical regression techniques with GP in the form of regression approaches expressed by symbols. The most significant applications of the EPR are applied during the extraction of the splendid mathematical model to fit the observed data sets.
http://www.hydroinformatics.it/index.php?option=com_docman&Itemid=105
I recommend Minitab for the regression model and Matlab for the 3D graphs.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS