Re: [R-sig-eco] Using the mra package in R...
Hi Sarah and all Thanks for those couple of tips. They did help to show that the data imported alright but that there was a distinct difference between my data set and the example data set for used to demonstrate the mra package. I have attached subset of my data as suggested. Its formatted as a tab-delimited text file with row names included (bird band numbers). I have tried importing with and without row names but this doesn't appear to make any difference. To import to R, I have been using the following process. brtb - read.table (file=filepath\\brtb.txt, header = TRUE, row.names = 1, sep = , colClasses=numeric) Inspecting the data using str(brtb) provides the following 'data.frame': 65 obs. of 19 variables: $ C1 : num 0 0 0 0 0 0 0 0 1 1 ... $ C2 : num 1 1 0 0 0 0 0 0 0 0 ... $ C3 : num 1 0 1 1 1 1 1 1 0 0 ... $ C4 : num 0 0 0 0 0 0 0 0 0 0 ... $ C5 : num 0 0 0 0 1 0 0 0 0 0 ... $ C6 : num 0 0 0 0 0 0 0 0 0 0 ... $ C7 : num 0 0 0 0 0 0 0 0 0 0 ... $ C8 : num 0 0 0 0 0 0 0 0 0 0 ... $ C9 : num 0 0 0 0 0 0 0 0 0 0 ... $ C10: num 0 0 0 0 0 0 0 0 1 0 ... $ C11: num 0 0 0 0 0 0 0 0 0 0 ... $ C12: num 0 0 0 0 0 0 0 0 0 0 ... $ C13: num 0 0 0 0 0 0 0 0 0 0 ... $ C14: num 0 0 0 0 0 0 0 0 0 0 ... $ C15: num 0 0 0 0 0 0 0 0 0 0 ... $ C16: num 0 0 0 0 0 0 0 0 0 0 ... $ C17: num 0 0 0 0 0 0 0 0 0 0 ... $ C18: num 0 0 0 0 0 0 0 0 0 0 ... $ C19: num 0 0 0 0 0 0 0 0 0 0 ... The comparative example I am looking at from the mra package is the dipper.histories data. Inspecting this with str(dipper.histories) provides the following num [1:294, 1:7] 1 1 1 1 1 1 1 1 1 1 ... - attr(*, dimnames)=List of 2 ..$ : chr [1:294] 1 2 3 4 ... ..$ : chr [1:7] h1 h2 h3 h4 ... There is an obvious difference in the returned output from the str() function. Inspection of both data frames shows they are formatted in a similar way, or at least appear to be. With respect to the function I am attempting to run in mra, I have run the examples routines as described in the mra manual using the supplied data. Results as described. In attempting this on any of my own data as attached I get an error message. The function I am trying to run is as follows brtb.cjs = F.cjs.estim( ~x2+x3+x4+x5+x6+x7+x8+x9+x10+x11+x12+x13+x14+x15+x16+x17+x18, ~x1+x2+x3+x4+x5+x6+x7+x8+x9+x10+x11+x12+x13+x14+x15+x16+x17, brtb) Each time I get the following error Error in F.cjs.estim(~x2 + x3 + x4 + x5 + x6 + x7 + x8 + x9 + x10 + x11 + : Capture histories must consist of 0's, 1's, and 2's only. I have checked thru the data-set to make sure the data matches these requirements, that there are no additional characters in the data-set, etc. and keep getting the same error. Have tried a different subsets of the data, different import options, etc. but still no success. Am assuming the error lies with the format of the data imported to R but cannot work out where or how to adjust to match the example. Any suggestions or advice on this one would be greatly appreciated. Cheers Adam... On 25 August 2013 23:33, Sarah Goslee sarah.gos...@gmail.com wrote: Hi, The first thing to do is str(yourdata) to double-check that your data import worked as expected. Particularly check that your data column isn't actually of mode factor. If that doesn't give you a lead on solving the problem, posting a reproducible example here with a subset of your data provided using dput() and the actual code you used will get you concrete answers, rather than random guesses. Sarah On Sunday, August 25, 2013, Adam Fawcett wrote: Have been trying to analyse life history data collected from a long term mark-recapture study of birds using the package mra. While the analysis works fine on the supplied example dataset, I keep getting error messages when running my data thru the F.cjs.estim function. Error message states that the data for this analysis must be either 1, 2 or0's. Have triple checked my data and its all 1 0's as required. Problems is likely to be an issue with either the read.table function to import my data (all correctly formatted and transformed ready for the analysis, possibly adding a character I have been able to remove, or with the mra package itself. Does anyone have experience using mra that could lend some advice. Alternatively, has anyone struck this sort of problem before that could suggest how to correct the problem. - -- Sarah Goslee http://www.stringpage.com http://www.sarahgoslee.com http://www.functionaldiversity.org -- *Adam Fawcett **Lecturer (GSE854 and GSE856) *Department of Environment and Geography Macquarie University NSW 2109 Australia M: + 61 427 929 554 http://mq.edu.aumq.edu.au CRICOS Provider Number 2J Please consider the environment before printing this email. This email (including all attachments) is confidential. It may be subject to legal professional privilege and/or protected by copyright. If you receive it in error do not use it or
Re: [R-sig-eco] Using the mra package in R...
Try coercing your data to matrix. You currently have a data.frame, which might matter to the package functions (I'm not familiar enough with the functions to be able to tell from the top of my head). as.matrix(my.data) A side note, here's how you would write a model that includes all (remaining) variables in your data: F.cjs.estim(~ ., data = ...) Notice the ., which means all available. Cheers, Roman On Mon, Aug 26, 2013 at 8:01 AM, Adam Fawcett adam.fawc...@mq.edu.auwrote: Hi Sarah and all Thanks for those couple of tips. They did help to show that the data imported alright but that there was a distinct difference between my data set and the example data set for used to demonstrate the mra package. I have attached subset of my data as suggested. Its formatted as a tab-delimited text file with row names included (bird band numbers). I have tried importing with and without row names but this doesn't appear to make any difference. To import to R, I have been using the following process. brtb - read.table (file=filepath\\brtb.txt, header = TRUE, row.names = 1, sep = , colClasses=numeric) Inspecting the data using str(brtb) provides the following 'data.frame': 65 obs. of 19 variables: $ C1 : num 0 0 0 0 0 0 0 0 1 1 ... $ C2 : num 1 1 0 0 0 0 0 0 0 0 ... $ C3 : num 1 0 1 1 1 1 1 1 0 0 ... $ C4 : num 0 0 0 0 0 0 0 0 0 0 ... $ C5 : num 0 0 0 0 1 0 0 0 0 0 ... $ C6 : num 0 0 0 0 0 0 0 0 0 0 ... $ C7 : num 0 0 0 0 0 0 0 0 0 0 ... $ C8 : num 0 0 0 0 0 0 0 0 0 0 ... $ C9 : num 0 0 0 0 0 0 0 0 0 0 ... $ C10: num 0 0 0 0 0 0 0 0 1 0 ... $ C11: num 0 0 0 0 0 0 0 0 0 0 ... $ C12: num 0 0 0 0 0 0 0 0 0 0 ... $ C13: num 0 0 0 0 0 0 0 0 0 0 ... $ C14: num 0 0 0 0 0 0 0 0 0 0 ... $ C15: num 0 0 0 0 0 0 0 0 0 0 ... $ C16: num 0 0 0 0 0 0 0 0 0 0 ... $ C17: num 0 0 0 0 0 0 0 0 0 0 ... $ C18: num 0 0 0 0 0 0 0 0 0 0 ... $ C19: num 0 0 0 0 0 0 0 0 0 0 ... The comparative example I am looking at from the mra package is the dipper.histories data. Inspecting this with str(dipper.histories) provides the following num [1:294, 1:7] 1 1 1 1 1 1 1 1 1 1 ... - attr(*, dimnames)=List of 2 ..$ : chr [1:294] 1 2 3 4 ... ..$ : chr [1:7] h1 h2 h3 h4 ... There is an obvious difference in the returned output from the str() function. Inspection of both data frames shows they are formatted in a similar way, or at least appear to be. With respect to the function I am attempting to run in mra, I have run the examples routines as described in the mra manual using the supplied data. Results as described. In attempting this on any of my own data as attached I get an error message. The function I am trying to run is as follows brtb.cjs = F.cjs.estim( ~x2+x3+x4+x5+x6+x7+x8+x9+x10+x11+x12+x13+x14+x15+x16+x17+x18, ~x1+x2+x3+x4+x5+x6+x7+x8+x9+x10+x11+x12+x13+x14+x15+x16+x17, brtb) Each time I get the following error Error in F.cjs.estim(~x2 + x3 + x4 + x5 + x6 + x7 + x8 + x9 + x10 + x11 + : Capture histories must consist of 0's, 1's, and 2's only. I have checked thru the data-set to make sure the data matches these requirements, that there are no additional characters in the data-set, etc. and keep getting the same error. Have tried a different subsets of the data, different import options, etc. but still no success. Am assuming the error lies with the format of the data imported to R but cannot work out where or how to adjust to match the example. Any suggestions or advice on this one would be greatly appreciated. Cheers Adam... On 25 August 2013 23:33, Sarah Goslee sarah.gos...@gmail.com wrote: Hi, The first thing to do is str(yourdata) to double-check that your data import worked as expected. Particularly check that your data column isn't actually of mode factor. If that doesn't give you a lead on solving the problem, posting a reproducible example here with a subset of your data provided using dput() and the actual code you used will get you concrete answers, rather than random guesses. Sarah On Sunday, August 25, 2013, Adam Fawcett wrote: Have been trying to analyse life history data collected from a long term mark-recapture study of birds using the package mra. While the analysis works fine on the supplied example dataset, I keep getting error messages when running my data thru the F.cjs.estim function. Error message states that the data for this analysis must be either 1, 2 or0's. Have triple checked my data and its all 1 0's as required. Problems is likely to be an issue with either the read.table function to import my data (all correctly formatted and transformed ready for the analysis, possibly adding a character I have been able to remove, or with the mra package itself. Does anyone have experience using mra that could lend some advice. Alternatively, has anyone struck this sort of problem before that could suggest how to correct the
Re: [R-sig-eco] Using the mra package in R...
Hi Roman and all That certainly helped a lot thanks. Data is now in the correct format and comparable to the example data set. Next problem that cropped up - I ran the F.cjs.estim function and got the following error brtb.cjs = F.cjs.estim( capture=~x2+x3+x4+x5+x6+x7+x8+x9+x10+x11+x12+x13+x14+x15+x16+x17+x18, survival=~x1+x2+x3+x4+x5+x6+x7+x8+x9+x10+x11+x12+x13+x14+x15+x16+x17, brtb.mx) Error in model.frame.default(formula = ~x1 + x2 + x3 + x4 + x5 + x6 + : variable lengths differ (found for 'x8') This has certainly thrown me. Ran traceback() to try and determine the problem which resulted with traceback()7: model.frame.default(formula = ~x1 + x2 + x3 + x4 + x5 + x6 + x7 + x8 + x9 + x10 + x11 + x12 + x13 + x14 + x15 + x16 + x17, drop.unused.levels = TRUE) 6: model.frame(formula = ~x1 + x2 + x3 + x4 + x5 + x6 + x7 + x8 + x9 + x10 + x11 + x12 + x13 + x14 + x15 + x16 + x17, drop.unused.levels = TRUE) 5: eval(expr, envir, enclos) 4: eval(mf) 3: F.3d.model.matrix(as.formula(survival), nan, ns) 2: F.cr.model.matrix(capture, survival, nan, ns) 1: F.cjs.estim(~x2 + x3 + x4 + x5 + x6 + x7 + x8 + x9 + x10 + x11 + x12 + x13 + x14 + x15 + x16 + x17 + x18, ~x1 + x2 + x3 + x4 + x5 + x6 + x7 + x8 + x9 + x10 + x11 + x12 + x13 + x14 + x15 + x16 + x17, brtb.mx) Suggestions and comments appreciated. Many thanks Adam... On 26 August 2013 16:10, Roman Luštrik roman.lust...@gmail.com wrote: Try coercing your data to matrix. You currently have a data.frame, which might matter to the package functions (I'm not familiar enough with the functions to be able to tell from the top of my head). as.matrix(my.data) A side note, here's how you would write a model that includes all (remaining) variables in your data: F.cjs.estim(~ ., data = ...) Notice the ., which means all available. Cheers, Roman On Mon, Aug 26, 2013 at 8:01 AM, Adam Fawcett adam.fawc...@mq.edu.auwrote: Hi Sarah and all Thanks for those couple of tips. They did help to show that the data imported alright but that there was a distinct difference between my data set and the example data set for used to demonstrate the mra package. I have attached subset of my data as suggested. Its formatted as a tab-delimited text file with row names included (bird band numbers). I have tried importing with and without row names but this doesn't appear to make any difference. To import to R, I have been using the following process. brtb - read.table (file=filepath\\brtb.txt, header = TRUE, row.names = 1, sep = , colClasses=numeric) Inspecting the data using str(brtb) provides the following 'data.frame': 65 obs. of 19 variables: $ C1 : num 0 0 0 0 0 0 0 0 1 1 ... $ C2 : num 1 1 0 0 0 0 0 0 0 0 ... $ C3 : num 1 0 1 1 1 1 1 1 0 0 ... $ C4 : num 0 0 0 0 0 0 0 0 0 0 ... $ C5 : num 0 0 0 0 1 0 0 0 0 0 ... $ C6 : num 0 0 0 0 0 0 0 0 0 0 ... $ C7 : num 0 0 0 0 0 0 0 0 0 0 ... $ C8 : num 0 0 0 0 0 0 0 0 0 0 ... $ C9 : num 0 0 0 0 0 0 0 0 0 0 ... $ C10: num 0 0 0 0 0 0 0 0 1 0 ... $ C11: num 0 0 0 0 0 0 0 0 0 0 ... $ C12: num 0 0 0 0 0 0 0 0 0 0 ... $ C13: num 0 0 0 0 0 0 0 0 0 0 ... $ C14: num 0 0 0 0 0 0 0 0 0 0 ... $ C15: num 0 0 0 0 0 0 0 0 0 0 ... $ C16: num 0 0 0 0 0 0 0 0 0 0 ... $ C17: num 0 0 0 0 0 0 0 0 0 0 ... $ C18: num 0 0 0 0 0 0 0 0 0 0 ... $ C19: num 0 0 0 0 0 0 0 0 0 0 ... The comparative example I am looking at from the mra package is the dipper.histories data. Inspecting this with str(dipper.histories) provides the following num [1:294, 1:7] 1 1 1 1 1 1 1 1 1 1 ... - attr(*, dimnames)=List of 2 ..$ : chr [1:294] 1 2 3 4 ... ..$ : chr [1:7] h1 h2 h3 h4 ... There is an obvious difference in the returned output from the str() function. Inspection of both data frames shows they are formatted in a similar way, or at least appear to be. With respect to the function I am attempting to run in mra, I have run the examples routines as described in the mra manual using the supplied data. Results as described. In attempting this on any of my own data as attached I get an error message. The function I am trying to run is as follows brtb.cjs = F.cjs.estim( ~x2+x3+x4+x5+x6+x7+x8+x9+x10+x11+x12+x13+x14+x15+x16+x17+x18, ~x1+x2+x3+x4+x5+x6+x7+x8+x9+x10+x11+x12+x13+x14+x15+x16+x17, brtb) Each time I get the following error Error in F.cjs.estim(~x2 + x3 + x4 + x5 + x6 + x7 + x8 + x9 + x10 + x11 + : Capture histories must consist of 0's, 1's, and 2's only. I have checked thru the data-set to make sure the data matches these requirements, that there are no additional characters in the data-set, etc. and keep getting the same error. Have tried a different subsets of the data, different import options, etc. but still no success. Am assuming the error lies with the format of the data imported to R but cannot work out where or how to adjust to match the example. Any suggestions or advice on
