BorlandTalk.com Borland discussion newsgroups   Archives   FAQ   Search   Memberlist   Usergroups   Register   Profile   Log in to check your private messages   Log in  Regression analysis -- how-to?          BorlandTalk.com Forum Index -> comp.lang.pascal.delphi.misc View previous topic :: View next topic   Author Message korax1214@mailandnews.co. Guest Posted: Sun Jan 09, 2005 8:11 pm    Post subject: Regression analysis -- how-to? (Yes, I know that sci.math would be more appropriate for this thread -- were it not for the fact that, the last time I asked an algorithm question there, the newsgroup's troll replied with a post which started "there's no such thing as rectangular or polar complex numbers, there are only complex numbers" (which may be true of *pure* maths[1] (although I won't believe this until I get confirmation from someone intelligent) but is definitely *not* true of *applied* maths (e.g. algorithms) where the distinction between rectangular and polar forms is very real (pun not intended) and quite important. For instance, the algorithm for multiplying two rectangular complex numbers is very different from the algorithm for multiplying two polar complex numbers, just as the algorithm for multiplying 1/2 by 3/4 is very different from the algorithm for multiplying 0.5 by 0.75, even though the same pair of fractions are involved in both cases), and went rapidly downhill from there. (The idiot in question obviously not only doesn't know what "twit" means, at least not in British English which is what I speak and write (it's *not* an insult), but from his reply to my reply to him, he obviously doesn't even know what "fraction" means.) Hence I am posting this to a group where I am likely to get *some* intelligent replies -- besides, I already tried sci.math with this question, and it appears that nobody there is capable of answering it.) Could anyone give me one or more algorithms for performing regression analysis, calculating the coefficient of correlation of a dataset, or any related topic of interest? Or give me a pointer as to where (preferably online) I could find this information? I know that there are at least six types of regression analysis (linear, quadratic, logarithmic, exponential, power and inverse) and at least two types of linear correlation coefficient (rank and product moment, IIRC), but I'd like to know the details... I know how to calculate the arithmetic mean and the product/population standard deviations of a dataset, which I think are at least 70% of what I need to do linear regression analysis, but I have no idea how to go beyond this. [1] Please don't tell me that the "correct" spelling is "math" (British English, remember?); "mathematics" is a plural-only word, so one can't have only one math, any more than one can have only one applau. :-) Back to top Tom de Neef Guest Posted: Sun Jan 09, 2005 8:19 pm    Post subject: Re: Regression analysis -- how-to? <korax1214 (AT) mailandnews (DOT) co.uk> schreef in bericht news:1105301500.658864.275730 (AT) c13g2000cwb (DOT) googlegroups.com... Quote: Could anyone give me one or more algorithms for performing regression analysis, calculating the coefficient of correlation of a dataset, or any related topic of interest? Or give me a pointer as to where (preferably online) I could find this information? When I google for "regression analysis delphi freeware" I get more than 7000 references and there are certainly some worthwhile ones amongst the first ten. Tom Back to top Maarten Wiltink Guest Posted: Sun Jan 09, 2005 9:51 pm    Post subject: Re: Regression analysis -- how-to? <korax1214 (AT) mailandnews (DOT) co.uk> wrote Quote: (Yes, I know that sci.math would be more appropriate for this thread -- were it not for the fact that, the last time I asked an algorithm question there, the newsgroup's troll replied with a post which started "there's no such thing as rectangular or polar complex numbers, there are only complex numbers" (which may be true of *pure* maths[1] (although I won't believe this until I get confirmation from someone intelligent) but is definitely *not* true of *applied* maths (e.g. algorithms) where the distinction between rectangular and polar forms is very real (pun not intended) and quite important. For instance, the algorithm for multiplying two rectangular complex numbers is very different from the algorithm for multiplying two polar complex numbers, just as the algorithm for multiplying 1/2 by 3/4 is very different from the algorithm for multiplying 0.5 by 0.75, even though the same pair of fractions are involved in both cases), and went rapidly downhill from there. (The idiot in question obviously not only doesn't know what "twit" means, at least not in British English which is what I speak and write (it's *not* an insult), but from his reply to my reply to him, he obviously doesn't even know what "fraction" means.) Hence I am posting this to a group where I am likely to get *some* intelligent replies -- besides, I already tried sci.math with this question, and it appears that nobody there is capable of answering it.) I tried to trim but failed. The troll has a point that being polar or Carthesian is not a property of the number, but of its notation. Every complex number can be expressed both as (x+i*y) for some x and y, and as r*e^(i*phi) for some r and phi. Several on-line dictionaries accessible through dictionary.com seem to agree that "twit" _is_ an insult. They also seem to speak "slang" better than British English, which is quite common on the Internet. It would also seem that 0.5 and 0.75 are fractions only in the sense of being between zero and one; they're not in the notation of one integer divided into another. As for intelligent replies, I think Tom gave one. Google knows everything about everything; incidentally MathWorld knows more than I do about math. Groetjes, Maarten Wiltink Back to top Mark Vaughan Guest Posted: Mon Jan 10, 2005 2:05 am    Post subject: Re: Regression analysis -- how-to? [email]korax1214 (AT) mailandnews (DOT) co.uk[/email] wrote in news:1105301500.658864.275730 @c13g2000cwb.googlegroups.com: Quote: Could anyone give me one or more algorithms for performing regression analysis, calculating the coefficient of correlation of a dataset, or any related topic of interest? Or give me a pointer as to where (preferably online) I could find this information? for code, browse the links at the URL in my sig for algorithms, books are better. the first two books listed here contain some useful info on regression techniques, with code examples in plain vanilla pascal: http://www-rab.larc.nasa.gov/nmp/fNMPbook.htm -- Mark Vaughan ____________ Visit the Numerical Methods in Pascal web page at http://www-rab.larc.nasa.gov/nmp/fNMPhome.htm Back to top korax1214@mailandnews.co. Guest Posted: Mon Jan 10, 2005 7:13 pm    Post subject: Re: Regression analysis -- how-to? Maarten Wiltink wrote: Quote: I tried to trim but failed. The troll has a point that being polar or Carthesian is not a property of the number, but of its notation. Every complex number can be expressed both as (x+i*y) for some x and y, and as r*e^(i*phi) for some r and phi. But that doesn't mean that there's "no such thing". As I already said, in applied mathematics (especially algorithms) the concrete form of the data *is* (all-)important. (Incidentally, I was taught to use "theta" rather than "phi" in the above expression, but that's probably just another cultural difference; it doesn't mean that either of us are "wrong".) Quote: Several on-line dictionaries accessible through dictionary.com seem to agree that "twit" _is_ an insult. Not in British English, it isn't (I've been speaking and writing that dialect for nearly 50 years); maybe you should find a better online dictionary. If you do a web/Usenet search (for robert.bak, korax1214 or robert@fm) you will find several posts where I call *myself* a twit. :-) Quote: It would also seem that 0.5 and 0.75 are fractions only in the sense of being between zero and one; they're not in the notation of one integer divided into another. But "one integer divided into another" is not a fraction in the general sense; it's specifically a *rational* (a.k.a. solidus) fraction. According to the Oxford English Dictionary, a fraction generally is "any number which is not an integer" (and even that definition may be too narrow, since of course the integers can be considered a special case of the reals, just as the reals are a subset of the complex numbers). Quote: As for intelligent replies, I think Tom gave one. Well, it's one more than I got from sci.trollheim. Back to top Maarten Wiltink Guest Posted: Mon Jan 10, 2005 8:23 pm    Post subject: Re: Regression analysis -- how-to? <korax1214 (AT) mailandnews (DOT) co.uk> wrote Quote: Maarten Wiltink wrote: [...] The troll has a point that being polar or Carthesian is not a property of the number, but of its notation. ... But that doesn't mean that there's "no such thing". As I already said, in applied mathematics (especially algorithms) the concrete form of the data *is* (all-)important. Okay, _you_ have a point there. But, as we pedants are fond of saying, that's just an implementation issue. Quote: (Incidentally, I was taught to use "theta" rather than "phi" in the above expression, but that's probably just another cultural difference; it doesn't mean that either of us are "wrong".) I've seen that. I think it was the manual for my calculator. But in university, it was all phi IIRC. Somehow, alpha, beta, gamma, phi, chi, and psi are angles; kappa, lambda, mu, nu, and ksi are linear parameters. Theta somehow doesn't have the "angle nature" right off. It doesn't clash but it simply wouldn't be my first pick. Quote: Several on-line dictionaries accessible through dictionary.com seem to agree that "twit" _is_ an insult. Not in British English, it isn't (I've been speaking and writing that dialect for nearly 50 years); maybe you should find a better online dictionary. I only know the one. I'm still awaiting standardisation of International English; until then I make do. Any dialect of English is better than Portuguese - although I _have_ answered the odd message in it. Quote: If you do a web/Usenet search (for robert.bak, korax1214 or robert@fm) you will find several posts where I call *myself* a twit. Oh yeah, like that proves anything. Don't we all? Hey, I thought "twit" wasn't an insult? (-: [...] Quote: According to the Oxford English Dictionary, a fraction generally is "any number which is not an integer" ... Too bad I don't own one. I have a Van Dale Groot Woordenboek der Nederlandse Taal, though, if you ever thirst for the meaning of "torntoffel"? Groetjes, Maarten Wiltink Back to top Tom de Neef Guest Posted: Mon Jan 10, 2005 10:32 pm    Post subject: Re: Regression analysis -- how-to? "Maarten Wiltink" <maarten (AT) kittensandcats (DOT) net> schreef Quote: [...] According to the Oxford English Dictionary, a fraction generally is "any number which is not an integer" ... Too bad I don't own one. I have a Van Dale Groot Woordenboek der Nederlandse Taal, though, if you ever thirst for the meaning of "torntoffel"? "Lastig of raar vrouwspersoon" (difficult or weird female). Tom Back to top Joseph McDonnell Guest Posted: Tue Jan 11, 2005 9:18 am    Post subject: Re: Regression analysis -- how-to? [email]korax1214 (AT) mailandnews (DOT) co.uk[/email] wrote: Quote: (Yes, I know that sci.math would be more appropriate for this thread -- were it not for the fact that, the last time I asked an algorithm question there, the newsgroup's troll replied with a post which started "there's no such thing as rectangular or polar complex numbers, there are only complex numbers" (which may be true of *pure* maths[1] (although I won't believe this until I get confirmation from someone intelligent) but is definitely *not* true of *applied* maths (e.g. algorithms) where the distinction between rectangular and polar forms is very real (pun not intended) and quite important. For instance, the algorithm for multiplying two rectangular complex numbers is very different from the algorithm for multiplying two polar complex numbers, just as the algorithm for multiplying 1/2 by 3/4 is very different from the algorithm for multiplying 0.5 by 0.75, even though the same pair of fractions are involved in both cases), and went rapidly downhill from there. (The idiot in question obviously not only doesn't know what "twit" means, at least not in British English which is what I speak and write (it's *not* an insult), but from his reply to my reply to him, he obviously doesn't even know what "fraction" means.) Hence I am posting this to a group where I am likely to get *some* intelligent replies -- besides, I already tried sci.math with this question, and it appears that nobody there is capable of answering it.) You've forgotten Monty Python's Upper Class Twit of the Year Contest. Quote: Could anyone give me one or more algorithms for performing regression analysis, calculating the coefficient of correlation of a dataset, or any related topic of interest? Or give me a pointer as to where (preferably online) I could find this information? May I ask why you want to do this? Do you want to learn about regression? Then I suggest you consult a few books. There are lots of books which will provide you with theory, which, for linear regression, is fairly simple but is much less so for other forms of regression. Trying to deduce what's going on from code would be difficult and difficult to generalise to a new situation. Quote: I know that there are at least six types of regression analysis (linear, quadratic, logarithmic, exponential, power and inverse) and at least two types of linear correlation coefficient (rank and product moment, IIRC), but I'd like to know the details... I suspect what you mean by 'quadratic' is the inclusion of an x^2 term as well as an x term. If so, you're still talking about linear regression ('linear' here means linear in the regression parameters, not in the x's). This might also be the case for the other forms you mention or you could be talking about different 'links' e.g. y = log(a + b*x). The theory for these kinds of models is a little more difficult, but you seem to have a maths (note the spelling) background. I'd recommend Annette Dobson's "Introduction to Generalized Linear Models". Quote: I know how to calculate the arithmetic mean and the product/population standard deviations of a dataset, which I think are at least 70% of what I need to do linear regression analysis, but I have no idea how to go beyond this. Basically you also need to calculate the correlation coefficients between the variables and you have all you need. Quote: [1] Please don't tell me that the "correct" spelling is "math" (British English, remember?); "mathematics" is a plural-only word, so one can't have only one math, any more than one can have only one applau. Even worse, them Yanks say "TOM-MAY-TOE." :-) Regards Joseph Back to top korax1214@mailandnews.co. Guest Posted: Wed Jan 12, 2005 5:37 pm    Post subject: Re: Regression analysis -- how-to? Joseph McDonnell wrote: Quote: korax1214 (AT) mailandnews (DOT) co.uk wrote: I suspect what you mean by 'quadratic' is the inclusion of an x^2 term as well as an x term. If so, you're still talking about linear regression ('linear' here means linear in the regression parameters, not in the x's). Well, my calculator offers "quadratic regression analysis" in which the data are assumed to conform (or to closely conform) to a parabola (y=a+bx+cx^2), although the manual also says that it can't calculate a coefficient of correlation for this type (it doesn't say whether this is because it's not possible, or because it's not feasible on a calculator), so (in this case at least) quadratic regression doesn't seem to be derived from linear, as the other four are (on this calculator at least). Quote: ... you seem to have a maths (note the spelling) background. I'd recommend Annette Dobson's "Introduction to Generalized Linear Models". Thanks. (I've saved this thread, apart from this reply of course, so I'll look up this book ASAP.) Back to top Alf Christophersen Guest Posted: Wed Feb 09, 2005 11:27 pm    Post subject: Re: Regression analysis -- how-to? On 12 Jan 2005 09:37:23 -0800, [email]korax1214 (AT) mailandnews (DOT) co.uk[/email] wrote: Quote: Well, my calculator offers "quadratic regression analysis" in which the data are assumed to conform (or to closely conform) to a parabola (y=a+bx+cx^2), although the manual also says that it can't calculate a coefficient of correlation for this type (it doesn't say whether this The opposite of linear regression is not quadratic or cubic regression analysis, but nonlinear regression, like y = a + bx +cxy To take the simplest case :-) Back to top korax1214@mailandnews.co. Guest Posted: Fri Feb 11, 2005 9:28 pm    Post subject: Re: Regression analysis -- how-to? Alf Christophersen wrote: Quote: On 12 Jan 2005 09:37:23 -0800, [email]korax1214 (AT) mailandnews (DOT) co.uk[/email] wrote: Well, my calculator offers "quadratic regression analysis" in which the data are assumed to conform (or to closely conform) to a parabola (y=a+bx+cx^2), although the manual also says that it can't calculate a coefficient of correlation for this type (it doesn't say whether this The opposite of linear regression is not quadratic or cubic regression analysis But who said anything about the opposite of linear regression analysis? I didn't, in the passage you quoted or elsewhere -- I'm asking about quadratic regression analysis... Back to top Alf Christophersen Guest Posted: Fri Feb 11, 2005 11:09 pm    Post subject: Re: Regression analysis -- how-to? On 11 Feb 2005 13:28:15 -0800, [email]korax1214 (AT) mailandnews (DOT) co.uk[/email] wrote: Quote: (y=a+bx+cx^2), although the manual also says that it can't calculate But who said anything about the opposite of linear regression analysis? I didn't, in the passage you quoted or elsewhere -- I'm asking about quadratic regression analysis... That is the linear regression to the function y=a+bx+cx2 :-) It is always a linear regresson as long as y is on the left side and depends only on x, or x and z etc. If y is on both side, it ain't not any longer a linear regression. Cubic regression is also a linear regression, but need more sums to be made :-) Just enlarge the matrix of constant, x-value, x^2-values of each dataset with x^2 and evt. x^^3. Many years ago I did a lot of linear regressions to both such equations and hyperbolas like y = vx/(k+x) and more complex functions. Back to top Display posts from previous: All Posts1 Day7 Days2 Weeks1 Month3 Months6 Months1 Year Oldest FirstNewest First         BorlandTalk.com Forum Index -> comp.lang.pascal.delphi.misc All times are GMT Page 1 of 1   Jump to: Select a forum C++ Builder Newsgroups----------------C++ BuilderC++ Builder (ActiveX)C++ Builder (Announcements)C++ Builder (Command Line Tools)C++ Builder DatabasesC++ Builder Databases (ADO)C++ Builder Databases (Desktop)C++ Builder Databases (InterBase Express)C++ Builder Databases (Multi-Tier)C++ Builder Databases (SQL Servers)C++ Builder (Graphics)C++ Builder (IDE)C++ Builder (Internet)C++ Builder (Internet Socket)C++ Builder (Internet Web)C++ Builder (Jobs)C++ Builder (Language)C++ Builder (Language C++)C++ Builder (Compatibility with MFC)C++ Builder (Native API)C++ Builder (Non-Technical)C++ Builder (Students)C++ Builder (Third Party Tools)C++ Builder (Upgrade)C++ Builder (VCL)C++ Builder (VCL Components Usage)C++ Builder (VCL Components Development)C++ Builder (VCL Usage)C++ Builder (WinAPI)C++ Builder Installation Delphi Newsgroups----------------Delphi (General)Delphi ActiveX Controls UsageDelphi ActiveX Controls DevelopmentDelphi As400Delphi DatabasesDelphi Databases (ADO)Delphi Databases (dbExpress)Delphi Databases (Desktop)Delphi Databases (InterBase Express)Delphi Databases (Multi-Tier)Delphi Databases (SQL Servers)Delphi DeploymentDelphi GraphicsDelphi IDEDelphi InternationalizationDelphi InternetDelphi Internet Isapi-WebBrokerDelphi Internet WebSnapDelphi Internet WinsockDelphi JobsDelphi Language BASMDelphi Language ObjectPascalDelphi Model Driven Architecture (General)Delphi Model Driven Architecture (Third Party)Delphi MultimediaDelphi Native APIDelphi Non-TechnicalDelphi ObjectPascalDelphi OLE AutomationDelphi OO designDelphi OpenToolsAPIDelphi Reporting-ChartingDelphi RTLDelphi StudentsDelphi TeamSourceDelphi Thirdparty ToolsDelphi Thirdparty Tools (General)Delphi Thirdparty Tools (IntraWeb)Delphi Thirdparty Tools (RaveReports)Delphi UpgradeDelphi VCLDelphi VCL ComponentsDelphi VCL Components UsingDelphi VCL Components WritingDelphi WebServices SOAPDelphi WinAPIDelphi XMLDelphi Installationcomp.lang.pascal.delphi.components.misccomp.lang.pascal.delphi.components.usagecomp.lang.pascal.delphi.components.writingcomp.lang.pascal.delphi.databasescomp.lang.pascal.delphi.misccomp.sources.delphi JBuilder Newsgroups----------------JBuilderJBuilder AnnouncementsJBuilder Applet IssuesJBuilder As400JBuilder CompilerJBuilder CORBA/RMIJBuilder DatabasesJBuilder DB SwingJBuilder DebuggerJBuilder DeploymentJBuilder DocumentationJBuilder EnterpriseJBuilder HandheldJBuilder IDEJBuilder JavaJBuilder JavaAPIJBuilder Java LanguageJBuilder Java SwingJBuilder Java BeansJBuilder Java Beans UsageJBuilder Java Beans DevelopmentJBuilder JBCLJBuilder JDataStoreJBuilder JobsJBuilder LinuxJBuilder MacintoshJBuilder & MobileSetJBuilder MultiLingual AppsJBuilder Non-TechnicalJBuilder OpenToolsJBuilder Servlets (JSP)JBuilder StudentsJBuilder Team DevelopmentJBuilder Thirdparty ToolsJBuilder UpgradeJBuilder Installation InterBase Newsgroups----------------InterBaseInterBase Bug DiscussionInterBase GeneralInterBase IBConsoleInterBase OpenSourceInterBase ReplicationInterBase SQLInterBase ThirdpartyInterBase Installation Miscellaneous Borland Newsgroups----------------AppCenterAppServerEnteraVisiBroker  You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum Powered by phpBB © 2001, 2006 phpBB Group SEO toolkit © 2004-2006 webmedic.