Re: Error when fiiting AR residual structure with ASReml-R

From: Butler, David <David.Butler_at_DEEDI.QLD.GOV.AU>
Date: Tue, 19 Jan 2010 17:30:56 +1000

Hi Fernando, There are a couple of issues going on here. Firstly, your data is irregular in the sense that there are differing numbers of rows within columns at some sites, and secondly there is a misleading error message in one of the beta releases of V3, which unfortunately you must be running. Asreml-R checks the number of observations in each section by comparing the product of the number of unique occurrences of each dimension to the number of rows in the data frame for that section. In your example, Site 1 is "rectangular" with 432 rows in the data, and 6 unique values of the Column factor and 72 unique values of Row. In this case 6x72=432 and all is ok. However, for Site 2 there is an irregular number of Rows within Columns and a total of 775 rows in that section of the data frame. In this instance but there are 11 unique values for Column and 72 unique values for Row, so asreml is expecting 792 observations in the dataframe (for Site==2) but only finds 775. This is an error but unfortunately it has reported the total number of observations (2202) in the error message. To fit the separable ar structure you have defined, you will have to "pad" the data with missing values so that each site is "rectangular". Alternatively you could fit one of the two dimensional power models to the irregular data. You should not have to relabel the Row or Column factor levels. I would encourage you to get a later version of asreml-R that (hopefully) has many of the beta release problems fixed. Dave. from: ASReml users discussion group [mailto:ASREML-L_at_DPI.NSW.GOV.AU] On Behalf Of GRIGNOLA, FERNANDO E [AG/6042] Sent: Friday, 15 January 2010 5:05 AM To: ASREML-L_at_DPI.NSW.GOV.AU Subject: Error when fiiting AR residual structure with ASReml-R Hello, I'm trying to fit the following "classroom" MET model with a test dataset, but I get an error coming from the rcov. met.1.asr <- asreml(yld ~ Site, random =~ diag(Site):geno, rcov=~at(Site):ar1(Column):ar1(Row), data=mmData) Error in asreml.rdflt2(Y, Var[!which], section, lvl, dataFrame[which.row, : Error model specifies 792 observations, but data section has 2202 rows The data structure is fairly unbalanced and I have sorted the dataframe by Site, Column and Row (all defined as factors) as required by the package. 1 2 4 5 6 7 8 9 10 11 12 13 14 15 Site1 0 0 0 72 72 0 0 72 72 72 72 0 0 0 Site2 0 0 0 70 72 66 72 72 72 72 67 72 72 72 Site3 71 72 72 72 72 0 0 66 72 0 0 0 0 0 Site4 0 0 0 72 72 0 0 72 65 0 0 71 70 72 I wonder if Columns and Rows need to be renumbered in some way or I am specifying the error model incorrectly. As a general rule, do Columns and Rows need to be (re)numbered consecutively? I would not think so but ... Thanks for your assistance. Fernando This e-mail message may contain privileged and/or confidential information, and is intended to be received only by persons entitled to receive such information. If you have received this e-mail in error, please notify the sender immediately. Please delete it and all attachments from any servers, hard drives or any other media. Other use of this e-mail by you is strictly prohibited. All e-mails and attachments sent and received are subject to monitoring, reading and archival by Monsanto, including its subsidiaries. The recipient of this e-mail is solely responsible for checking for the presence of "Viruses" or other "Malware". Monsanto, along with its subsidiaries, accepts no liability for any damage caused by any such code transmitted by or accompanying this e-mail or any attachment. This message is intended for the addressee named and may contain confidential information. If you are not the intended recipient, please delete it and notify the sender. Views expressed in this message are those of the individual sender, and are not necessarily the views of their organisation. ********************************DISCLAIMER**************************** The information contained in the above e-mail message or messages (which includes any attachments) is confidential and may be legally privileged. It is intended only for the use of the person or entity to which it is addressed. If you are not the addressee any form of disclosure, copying, modification, distribution or any action taken or omitted in reliance on the information is unauthorised. Opinions contained in the message(s) do not necessarily reflect the opinions of the Queensland Government and its authorities. If you received this communication in error, please notify the sender immediately and delete it from your computer system network.
Received on Tue Jan 19 2010 - 17:30:56 EST

This webpage is part of the ASReml-l discussion list archives 2004-2010. More information on ASReml can be found at the VSN website. This discussion list is now deprecated - please use the VSN forum for discussion on ASReml. (These online archives were generated using the hypermail package.)