Hi all, I'm not familiar with R programming and I'm trying to reproduce a result from a paper. Basically, I have a dataset which I would like to model in terms of successive increments, i.e. (y denote empirical values of y) y_1 = y1, y_2 = y1 + delta1, y_3 = y1 + delta1 + delta2. ... y_m = y1 + sum_2^m delta j where delta_j donote successive increments in the y-values, i.e. delta j = y_j - y_(j-1). In order to estimate y-values, I'm assuming that delta j is approximately equal to kj**u, such that my regression model should be something like this: ^y_1 = a1 ^y_2 = a1 + k2**u ^y_3 = a1 + k2**u + k3**u ... ^y_m = a1 + k2**u + k3**u + ... + km**u or, generically ^yi = a1 + k * sum_j=2^i j**u and I need to fit a non-linear least-squares regression model to find the tripplet a1,k,u. I had a look to the gnm package, but I don't have the lesser idea how to formulate this problem to use this package. Can someone help me with that? cheers, Ronaldo
Hi all, I'm not familiar with R programming and I'm trying to reproduce a result from a paper. Basically, I have a dataset which I would like to model in terms of successive increments, i.e. (y denote empirical values of y) y_1 = y1, y_2 = y1 + delta1, y_3 = y1 + delta1 + delta2. ... y_m = y1 + sum_2^m delta j where delta_j donote successive increments in the y-values, i.e. delta j = y_j - y_(j-1). In order to estimate y-values, I'm assuming that delta j is approximately equal to kj**u, such that my regression model should be something like this: ^y_1 = a1 ^y_2 = a1 + k2**u ^y_3 = a1 + k2**u + k3**u ... ^y_m = a1 + k2**u + k3**u + ... + km**u or, generically ^yi = a1 + k * sum_j=2^i j**u and I need to fit a non-linear least-squares regression model to find the tripplet a1,k,u. I had a look to the gnm package, but I don't have the lesser idea how to formulate this problem to use this package. Can someone help me with that? cheers, Ronaldo
Not sure why you feel the need to use gnm here - are you working with non-normal data? From your description it would seem that nls is more appropriate, Heather Dr H Turner Research Fellow Dept. of Statistics The University of Warwick Coventry CV4 7AL Tel: 024 76575870 Fax: 024 76524532 Url: www.warwick.ac.uk/go/heatherturner ________________________________ From: Ronaldo Prati [mailto:prati@icmc.usp.br] Sent: Thu 14/12/2006 13:41 To: r-help@stat.math.ethz.ch Subject: [R] Model formula question Hi all, I'm not familiar with R programming and I'm trying to reproduce a result from a paper. Basically, I have a dataset which I would like to model in terms of successive increments, i.e. (y denote empirical values of y) y_1 = y1, y_2 = y1 + delta1, y_3 = y1 + delta1 + delta2. .. y_m = y1 + sum_2^m delta j where delta_j donote successive increments in the y-values, i.e. delta j = y_j - y_(j-1). In order to estimate y-values, I'm assuming that delta j is approximately equal to kj**u, such that my regression model should be something like this: ^y_1 = a1 ^y_2 = a1 + k2**u ^y_3 = a1 + k2**u + k3**u .. ^y_m = a1 + k2**u + k3**u + ... + km**u or, generically ^yi = a1 + k * sum_j=2^i j**u and I need to fit a non-linear least-squares regression model to find the tripplet a1,k,u. I had a look to the gnm package, but I don't have the lesser idea how to formulate this problem to use this package. Can someone help me with that? cheers, Ronaldo [[alternative HTML version deleted]]
[resend - hopefully HTML switched off this time] Not sure why you feel the need to use gnm here - are you working with non-normal data? From your description it would seem that nls is more appropriate, Heather Dr H Turner Research Fellow Dept. of Statistics The University of Warwick Coventry CV4 7AL Tel: 024 76575870 Fax: 024 76524532 Url: www.warwick.ac.uk/go/heatherturner -----Original Message----- From: Ronaldo Prati [mailto:prati@icmc.usp.br] Sent: Thu 14/12/2006 13:41 To: r-help@stat.math.ethz.ch Subject: [R] Model formula question Hi all, I'm not familiar with R programming and I'm trying to reproduce a result from a paper. Basically, I have a dataset which I would like to model in terms of successive increments, i.e. (y denote empirical values of y) y_1 = y1, y_2 = y1 + delta1, y_3 = y1 + delta1 + delta2. .. y_m = y1 + sum_2^m delta j where delta_j donote successive increments in the y-values, i.e. delta j = y_j - y_(j-1). In order to estimate y-values, I'm assuming that delta j is approximately equal to kj**u, such that my regression model should be something like this: ^y_1 = a1 ^y_2 = a1 + k2**u ^y_3 = a1 + k2**u + k3**u .. ^y_m = a1 + k2**u + k3**u + ... + km**u or, generically ^yi = a1 + k * sum_j=2^i j**u and I need to fit a non-linear least-squares regression model to find the tripplet a1,k,u. I had a look to the gnm package, but I don't have the lesser idea how to formulate this problem to use this package. Can someone help me with that? cheers, Ronaldo [[alternative HTML version deleted]]