similar to: Computing s^2 for non-negative least squares.

Thank you Brian Ripley, Gardar Johannesson, and Marcel Wolbers for your prompt and friendly help! I will share any further learnings as I move through these suggestions. -Bob Abugov Brian Ripley wrote: I just use optim() on the sum of squares with non-negativity constraints. That did not exist in 1999. Gardar Johannesson wrote: You can always just use the quadratic programing library in R

In July of 1999 Douglas Bates invited R users to implement an algorithm for non-negative least squares based on Bates and Wolf, 1984: Communications in Statistics, Part B 13:841-850. <http://www.ens.gu.edu.au/robertk/R/help/99b/0058.html> I'm wondering if anybody has implemented this or something similar so I won't have to reinvent the wheel. Thanks! Bob Abugov

Hello R experts, I have two questions related to the nnls library (http://www.inside-r.org/packages/cran/nnls), and more broadly to linear regression with positive coefficients. Sample code is below the Qs. Q1: Regular regression (with lm) gives me the significance of each variable. How do I get variable significance with nnls? If there's no ready function, any easy way to manually derive

Hi all, I am trying to do an inversion of electromagnetic data with non-negative least squares method (Tikhonov regularisation) and have got it programmed in S-Plus. However I am trying to move all my scripts from S-Plus to R. Is there an equivalent to nnls.fit in R? I think this can be done with pcls? Right? S-Plus script: A, L and data are matrices, lambda is a vector of possible lambda

Dear R-ians, I want to perform an linear unmixing of image pixels in fractions of pure endmembers. Therefore I need to perform a constrained linear least-squares problem that looks like : min || Cx - d || ? where sum(x) = 1. I have a 3x3 matrix C, containing the values for endmembers and I have a 3x1 column vector d (for every pixel in the image). In theory my x values should all be in the

I am using the non-linear least squares routine in "R" -- nls. I have a dataset where the nls routine outputs tight confidence intervals on the 2 parameters I am solving for. As a check on my results, I used the Python SciPy leastsq module on the same data set and it yields the same answer as "R" for the coefficients. However, what was somewhat surprising was the the

I would really appreciate it if someone can give suggestions on how to do spectra fitting in R using ordinary least square fitting and non-negativity constraints. The lm() function works well for ordinary least square fitting, but how to specify non-negativity constraints? It wouldn't make sense if the fitting coefficients coming out as negative in absorption spectra deconvolution. Thanks.

Hi I am looking for suitable packages in R that do regression analyses using least median squares method (or better). Additionally, I am also looking for packages that implement algorithms/methods for detecting outliers that can be discarded before doing the regression analyses. Although some websites refer to "lms" method under package "lps" in R, I am unable to find such a

Thanks for the response. I was not very clear in my original request. What I am asking is if in a non-linear estimation problem using nls(), as the condition number of the Hessian matrix becomes larger, will the t-values of one or more of the parameters being estimated in general become smaller in absolute value -- that is, are low t-values a sign of an ill-conditioned Hessian? Typical

I am using "nls" to fit dose-response curves but am not sure how to approach more robust regression in R to get around the problem of the my error showing increased variance with increasing dose. My understanding is that "rlm" or "lqs" would not be a good idea here. 'Fairly new to regression work, so apologies if I'm missing something obvious.

Hello, I am trying to optimise a nonlinear model to derive 'best-fit' parameter esimates using the genoud function. I have been using the genetic algorithm - gafit - in order to do this, but I am getting parameter estimates that do not always reach the global minimum. I am very keen to apply genoud to optimising this model to see if my results will improve, and also out of personal

How do you specify matrix location a[i,j] (or a[i-1,j], etc.) in a "for" loop? I am looking for a flexible method of indexing neighbors over a series of lags (1,2,3...) and I may wish to extend this method to 3D arrays. Example: Data matrix > fun [,1] [,2] [,3] [1,] 1 5 9 [2,] 2 6 10 [3,] 3 7 11 [4,] 4 8 12 For each element a[i,j] in

Hi, everyone, My question is: It's not every time that you can get a converged result from the nls function. Is there any solution for me to get a reasonable result? For example: x <- c(-0.06,-0.04,-0.025,-0.015,-0.005,0.005,0.015,0.025,0.04,0.06) y <- c(1866760,1457870,1314960,1250560,1184850,1144920,1158850,1199910,1263850,1452520) fitOup<- nls(y ~ constant + A*(x-MA)^4 +

Dear Mr Graves, Thank you very much for your response. Nobody else from this mailing list ventured to reply to me for the two weeks since I posted my question. "nlminb" and "optim" are just optimization procedures. What I need is not just optimization, but a nonlinear CURVE FITTING procedure. If there is some way to perform nonlinear curve fitting with the

Is there a function in R for constrained linear least squares? I used the matlab function LSQLIN: my aim is to obtain non-negative regression coefficients which sum 1. Thanks in advance, domenico vistocco ___________________________________ Yahoo! Mail: gratis 1GB per i messaggi e allegati da 10MB http://mail.yahoo.it

Hi, I am trying to use the nls() function to closely approximate a vector of values, colC and I'm running into trouble. I am not sure how if I am asking the program to do what I think its doing, because the same minimization in Excel's Solver does not run into problems. If anyone can tell me what is going wrong, and why I'm getting a singular convergence(7) error, please tell me. I

As John noted, there are different kinds of weights, and different terminology: * inverse-variance weights (accuracy weights) * case weights (frequencies, counts) * sampling weights (selection probability weights) I'll add: * inverse-variance weights, where var(y for observation) = 1/weight (as opposed to just being inversely proportional to the weight) * weights used as part of an

Does anyone can provide a code to implement least median squares regression in R (not using the lqs function or calling C functions)? Reason: teaching/learning purposes Thanks PM

I have a bayesian hierarchical normal regression model, in which the regression coefficients are nested, which I've wrapped into one regression framework, y = X %*% beta + e . I would like to run data through the model in a filter style (kalman filterish), updating regression coefficients at each step new data can be gathered. After the first filter step, I will need to be able to feed

Dear R Users, I would like to make a lesast squares regression similar to that what is done by the command "lm". But additionally, I would like to impose some restrictions: 1) The sum of all regression coefficients should be equal to 1. 2) Each coefficient should assume a value between 0 and 1. (inequality restrictions) Which command is the best to use in order to solve this problem