similar to: strange nlme augpred behaviour

Hi Well, yes I tried it about two weeks ago but my post did not get through as it still awaits moderator approval. I could check which column is offending but actually it is only minor nuisance, I can live with selection of columns before fitting a model. What seems to me strange is that both full dataset and only selected colums gave me identical fit results but only one works within augPred.

> On Aug 23, 2017, at 8:08 AM, PIKAL Petr <petr.pikal at precheza.cz> wrote: > > Hi > > Well, yes I tried it about two weeks ago but my post did not get through as it still awaits moderator approval. It got through just fine. It appeared on Aug 15. It just didn't get any replies. As I read your original question in this thread, it was not clear to me that you had

Better posted on r-sig-mixed-models , no? Cheers, Bert Bert Gunter "The trouble with having an open mind is that people keep coming along and sticking things into it." -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip ) On Wed, Aug 23, 2017 at 5:17 AM, PIKAL Petr <petr.pikal at precheza.cz> wrote: > Dear all > > I encountered strange

Hi R User, I am new in R and trying to create tables with selecting rows randomly (but with replacement) for each group but each group should have same number as original. Is it possible to create it using the following example data set? Your help is highly appreciated. dat1<-structure(list(RegionA = structure(c(1L, 1L, 2L, 3L, 3L, 4L, 5L, 5L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L),

I am attempting to graph 12 months of temperatures, delineate the months with a vline and place the names of the months at the top of the graph. So far I have gotten everything to work except the names, despite getting a similar graph to work yesterday the day before yesterday with Baptise A's help. Can anyone suggest what I am doing wrong. Data set is below code. Thanks. Code

Hi: I need some help with the legend. I got 14 samples(Muestreo) and I am trying to plot a smooth line for each sample. I am able to accomplish that but the problem is that the legend only displays every other sample. How can I force the legend to show all of my Muestreos? Thanks in advance. fish_ByMuestreo <- structure(list(data = structure(list(SampleDate = structure(c(3L, 3L, 3L, 3L,

Dear All, 1.I have been trying to aggregate my data but I have not been getting it correctly. In the data I want to convert hourly data into daily averages. Here is a sample of the data: Neph_no Date Time Temp_C Pressure_kPa RH 9 2014/03/28 10:00:00 38.4 95.9 29.7 9 2014/03/28 11:00:00 37.8 95.8 29.2 9 2014/03/28 12:00:00 36.7 95.8 35.1 9 2014/03/28 13:00:00 35.4 95.8 38.9 9

Hello everyone, I have a dataset with 5 colums (4 colums with thresholds of weather stations and one with month - data of 5 years). Now I would like to calculate the average for each month. I tried this unsuccessfully: lf.med <- sapply(LF[,1:4],mean,LF[,5]) Error in mean.default(X[[1L]], ...) : 'trim' must be numeric and have length 1 With lf.med <- by(LF[,1:4],LF[,5],mean)

Dear r-users, I would like to construct 3-day moving average for block maxima series. I tried this: bmthree <- lapply(split(dt, dt$Year), function(x) max(sapply(1:(nrow(x)-2), function(i) with(x, mean(Amount[i:(i+2)],na.rm=TRUE))))) bmthree and got the following output. $`1971` [1] 70.81667 $`1972` [1] 68.94553 $`1973` [1] 102.7236 $`1974` [1] 73.6625 $`1975` [1]

I had a problem annotating a graph last year ( see http://n4.nabble.com/Putting-names-on-a-ggplot-td907158.html#a907158 for the discussion) Stefan (smu) provided a solution using annotate(). However I apparently did not update the graph file and,now, when I go back to the thread and try to use Stefan's solution it does not seem to work although I am sure that it did then. The problem

Dear friends, I have a weekly time series which starts on Jan 4th, 2003 and ends on december 31st, 2016. I set up my ts object as follows: MyTseries <- ts(mydataset, start=2003, end=2016, frequency=52) MyModel <- auto.arima(MyTseries, d=1, D=1) MyModelForecast <- forecast (MyModel, h=12) Since my last observation was on december 31st, 2016 I expected my forecast date to start on

d <- structure(list(Site = structure(c(8L, 12L, 7L, 6L, 11L, 5L, 10L, 4L, 3L, 2L, 1L, 9L, 8L, 12L, 7L, 6L, 11L, 5L, 10L, 4L, 3L, 2L, 1L, 9L, 8L, 12L, 7L, 6L, 11L, 5L, 10L, 4L, 3L, 2L, 1L, 9L, 8L, 12L, 7L, 6L, 11L, 5L, 10L, 4L, 3L, 2L, 1L, 9L, 8L, 12L, 7L, 6L, 11L, 5L, 10L, 4L, 3L, 2L, 1L, 9L, 8L, 12L, 7L, 6L, 11L, 5L, 10L, 4L, 3L, 2L, 1L, 9L, 8L, 12L, 7L, 6L, 11L, 5L, 10L, 4L, 3L, 2L, 1L, 9L,

> On Dec 6, 2017, at 5:07 AM, Paul Bernal <paulbernal07 at gmail.com> wrote: > > Dear friends, > > I have a weekly time series which starts on Jan 4th, 2003 and ends on > december 31st, 2016. > > I set up my ts object as follows: > > MyTseries <- ts(mydataset, start=2003, end=2016, frequency=52) > > MyModel <- auto.arima(MyTseries, d=1, D=1)

Thank you very much David. As a matter of fact, I solved it by doing the following: MyTimeSeriesObj <- ts(MyData, freq=365.25/7, start=decimal_date(mdy("01-04-2003"))) After doing that adjustment, my forecasts dates started from 2017 on. Cheers, Paul 2017-12-06 12:03 GMT-05:00 David Winsemius <dwinsemius at comcast.net>: > > > On Dec 6, 2017, at 5:07 AM, Paul

I am trying to plot the following data using dotchart intersect.data<-structure(list(X = structure(c(1L, 3L, 8L, 9L, 10L, 11L, 12L, 13L, 14L, 15L, 2L, 4L, 5L, 6L, 7L), .Label = c("1-100", "1001-1100", "101-200", "1101-1200", "1201-1300", "1301-1400", "1401-1500", "201-300", "301-400", "401-500",

I think the subject says it all. I want to make a simple lattice plot, using xyplot with the argument type=c("l","a"). The problem then is that in the resulting plot it is difficult/impossible to see which plot corresponds to the average and which to the individual profiles. I triedthings like extra arguments lwd=c(1,3) or col=c("blue","red") hoping

I have a dataframe df with columns x, y, and height. I want to create a heatmap-like plot that creates a grid of x by y, and then color codes the grid depending on the value of height. Is there a ggplot2 object to do this? I'm able to easily do this in Excel with pivot tables and conditional formatting so I'm including an image that is close to the output I want. I want to be able to

Hi maybe there is another more elegant solution but something like this > idx <- 1:nrow(dat1) > lll <- split(idx, dat1$group) > dat1[unlist(lapply(lll, sample, rep=TRUE)),] gives you selected rows. You could use for cycle or save those data frames manually Cheers Petr > -----Original Message----- > From: R-help [mailto:r-help-bounces at r-project.org] On Behalf Of Marna

Hi I am very new to R, I am attempting to produce a monthly boxplot with the following fish thermal telemetry data: ID Temp Date.Time Month.Year Month Week Shortdate 1 1734 4.4140 04/05/2010 11:56 05,2010 May 19 04/05/2010 2 1734 4.1002 04/05/2010 12:06 05,2010 May 19 04/05/2010 3 1734 3.9433 04/05/2010 12:09 05,2010 May 19 04/05/2010 4 1734 3.6295

Hi, I am trying to plot the following data so that it can be visually represented well. I tried the dotchart but I felt it was too spread out. Then I tried the barplot which is good enough for me. Is there a way to give the labels for the y-axis as in the dot chart? Also, I feel the grey level is confusing, so is there options for designs within the bars? I cannot use color as the journal wants