Hello
Does TeeChart Pro v7 support standard linear regression in the form of y=mx+c?
If so please can you provide an example of this. Is it within your demo program?
Cheers
Paul
Linear Regression
-
- Site Admin
- Posts: 14730
- Joined: Mon Jun 09, 2003 4:00 am
- Location: Banyoles, Catalonia
- Contact:
Hi Paul,
You can do that with TCustomTeeFunction. There's an example of that at the features demo at All Features\Welcome!\Functions\Extended\Custom y=f(x).
You can do that with TCustomTeeFunction. There's an example of that at the features demo at All Features\Welcome!\Functions\Extended\Custom y=f(x).
Best Regards,
Narcís Calvet / Development & Support Steema Software Avinguda Montilivi 33, 17003 Girona, Catalonia Tel: 34 972 218 797 http://www.steema.com |
Instructions - How to post in this forum |
-
- Newbie
- Posts: 59
- Joined: Fri May 28, 2004 4:00 am
- Location: Sweden
Calculating regression i.e fitting parameters etc..
Hi,
Dont know if this helps, but I think you have to get yourself a component/unit which calculates a regression line, i.e fits parameters, calculates standard deviation for theese,t-values etc.. There are a variety of avaliable products commercial and free. Some also includes diagnostic tests example test for homoscedstic residuals variance etc. Beware however for many of theese avaliable component are not that good of finding best solutions for parameter fits when example mulitcollinearity is present within the independent variables.
I know for example that 1 of steemas employees och ex employee is involved with a company called dewresearch(www.dewresearch.com) which have a component for regression linear and non-linear. They also offer a good TMtvVector component for increased perfomance using arrays of different types.
Dont know if this helps you, but good luck, let me know how it goes, and if you have any problem, maybe I can assist you in some way.
Best Regards,
Johan Ingemansson
Dont know if this helps, but I think you have to get yourself a component/unit which calculates a regression line, i.e fits parameters, calculates standard deviation for theese,t-values etc.. There are a variety of avaliable products commercial and free. Some also includes diagnostic tests example test for homoscedstic residuals variance etc. Beware however for many of theese avaliable component are not that good of finding best solutions for parameter fits when example mulitcollinearity is present within the independent variables.
I know for example that 1 of steemas employees och ex employee is involved with a company called dewresearch(www.dewresearch.com) which have a component for regression linear and non-linear. They also offer a good TMtvVector component for increased perfomance using arrays of different types.
Dont know if this helps you, but good luck, let me know how it goes, and if you have any problem, maybe I can assist you in some way.
Best Regards,
Johan Ingemansson