An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20130321/cd489c00/attachment.pl>
contourplot
3 messages · Steven LeBlanc, Pascal Oettli, David L Carlson
Hi, What is the result of: contourplot(Y~X1+X2,data=pr2) HTH, Pascal Le 13/03/22 8:57, Steven LeBlanc a ?crit :
Greets, I'm using a data frame that looks like:
head(pr2)
X1 X2 X3 X4 Y fit res 1 44 33.2 5 30 41.2 39.22201 1.977991 2 43 33.8 4 41 31.7 38.48476 -6.784761 3 48 40.6 3 38 39.4 44.78278 -5.382783 4 52 39.2 7 48 57.5 51.48134 6.018656 5 71 45.5 11 53 74.8 68.25585 6.544153 6 44 37.5 9 65 59.8 53.27743 6.522569 Along with the command:
contourplot(Y~X1*X2,data=pr2)
But I get a blank plot. I thought it might be because the data were unsorted or sparse, so I made another ordered data frame as follows:
head(new)
X1 X2 X3 X4 resp 1 1 1 1 1 -10.810406 2 2 2 2 2 -7.657712 3 3 3 3 3 -4.505018 4 4 4 4 4 -1.352323 5 5 5 5 5 1.800371 6 6 6 6 6 4.953065 But the result is the same. Any idea why this does not work? Best Regards, Steven [[alternative HTML version deleted]]
______________________________________________ R-help at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Look carefully at the help file. Your x1 and x2 need to define a rectangular grid so that there is a value of y for each combination of different x1 and x2 values. When this is not the case, you don't get an error, just a blank plot. It is not clear from the little piece of data that you included if that is the case, but the blank plot suggests that you don't have a grid. You can created gridded data using expand.grid and predict.lm() to generate estimates of y for every combination of x1 and x2:
# Reproducible data set.seed(42) x1 <- rnorm(100, 60, 10) x2 <- rnorm(100, 37, 5) y <- x1+x2 pr <- data.frame(x1, x2, y) # Create grid that spans x1/x2 ranges # 50 values each ranging from min to max xg <- seq(min(x1), max(x2), length.out=50) yg <- seq(min(x2), max(x2), length.out=50) # Grid will have 50 x 50 = 2500 values xygrid <- expand.grid(xg, yg) # Create data frame from gridded data and compute y # from a linear regression of x1 and x2 on y lmmodel <- lm(y~x1+x2, pr) prgrid <- data.frame(x1=xygrid$Var1, x2=xygrid$Var2) prgrid$y <- predict(lmmodel, prgrid) # Draw contour plot contourplot(y~x1+x2, prgrid)
---------------------------------------------- David L Carlson Associate Professor of Anthropology Texas A&M University College Station, TX 77843-4352
-----Original Message----- From: r-help-bounces at r-project.org [mailto:r-help-bounces at r- project.org] On Behalf Of Pascal Oettli Sent: Thursday, March 21, 2013 11:41 PM To: Steven LeBlanc Cc: r-help at r-project.org Subject: Re: [R] contourplot Hi, What is the result of: contourplot(Y~X1+X2,data=pr2) HTH, Pascal Le 13/03/22 8:57, Steven LeBlanc a ?crit :
Greets, I'm using a data frame that looks like:
head(pr2)
X1 X2 X3 X4 Y fit res 1 44 33.2 5 30 41.2 39.22201 1.977991 2 43 33.8 4 41 31.7 38.48476 -6.784761 3 48 40.6 3 38 39.4 44.78278 -5.382783 4 52 39.2 7 48 57.5 51.48134 6.018656 5 71 45.5 11 53 74.8 68.25585 6.544153 6 44 37.5 9 65 59.8 53.27743 6.522569 Along with the command:
contourplot(Y~X1*X2,data=pr2)
But I get a blank plot. I thought it might be because the data were
unsorted or sparse, so I made another ordered data frame as follows:
head(new)
X1 X2 X3 X4 resp 1 1 1 1 1 -10.810406 2 2 2 2 2 -7.657712 3 3 3 3 3 -4.505018 4 4 4 4 4 -1.352323 5 5 5 5 5 1.800371 6 6 6 6 6 4.953065 But the result is the same. Any idea why this does not work? Best Regards, Steven [[alternative HTML version deleted]]
______________________________________________ R-help at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-
guide.html
and provide commented, minimal, self-contained, reproducible code.
______________________________________________ R-help at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting- guide.html and provide commented, minimal, self-contained, reproducible code.