Hi all Just wondering whether one can undertake Lp norm estimation (a type of regression analysis) in R? i.e. argmin S ( | y(i) - x(i)b | ^p ) where: * S is the summation over observation i= 1,2,...,n * y is a vector of n observations * x is an n by p matrix of explanatory variables * b is a p by 1 vector of beta coefficients and * p is a constant to be estimated such that p>= 1 Regards Allan [[alternative HTML version deleted]]
Allan - Brian Ripley's implementation of one of the more useful Lp norms is: library("lqs") help("lqs") This is very highly recommended for practical data analysis. - tom blackwell - u michigan medical school - ann arbor - On Thu, 18 Dec 2003, allan clark wrote:> Hi all > > Just wondering whether one can undertake Lp norm estimation (a type of > regression analysis) in R? > > i.e. > > argmin S ( | y(i) - x(i)b | ^p ) > > where: > > * S is the summation over observation i= 1,2,...,n > * y is a vector of n observations > * x is an n by p matrix of explanatory variables > * b is a p by 1 vector of beta coefficients and > * p is a constant to be estimated such that p>= 1 > > Regards > Allan >
Hi Allan, I think the function lmp() of the package normalp could be useful to you. Regards, Elio allan clark wrote:>Hi all > >Just wondering whether one can undertake Lp norm estimation (a type of >regression analysis) in R? > >i.e. > >argmin S ( | y(i) - x(i)b | ^p ) > >where: > > * S is the summation over observation i= 1,2,...,n > * y is a vector of n observations > * x is an n by p matrix of explanatory variables > * b is a p by 1 vector of beta coefficients and > * p is a constant to be estimated such that p>= 1 > >Regards >Allan > > [[alternative HTML version deleted]] > >______________________________________________ >R-help at stat.math.ethz.ch mailing list >https://www.stat.math.ethz.ch/mailman/listinfo/r-help > > >