[R] time series contains internal NAs error

Gabor Grothendieck ggrothendieck at gmail.com
Mon Jan 19 15:50:04 CET 2009


The statement to read in the data is missing from your post but
I suspect that you are representing the data as daily data so its filling
in 364 or 365 NA's between points.  Represent it as the annual data
that it is.  Try this:

Lines <- "Sunspots,Datefield
9.5,1/1/1900
2.7,1/1/1901
5,1/1/1902
24.4,1/1/1903
42,1/1/1904
63.5,1/1/1905
53.8,1/1/1906
62,1/1/1907
48.5,1/1/1908
43.9,1/1/1909
18.6,1/1/1910
5.7,1/1/1911
3.6,1/1/1912
1.4,1/1/1913
9.6,1/1/1914
47.4,1/1/1915
57.1,1/1/1916
103.9,1/1/1917
80.6,1/1/1918
63.6,1/1/1919
37.6,1/1/1920
26.1,1/1/1921
14.2,1/1/1922
5.8,1/1/1923
16.7,1/1/1924
44.3,1/1/1925
63.9,1/1/1926
69,1/1/1927
77.8,1/1/1928
64.9,1/1/1929
35.7,1/1/1930
21.2,1/1/1931
11.1,1/1/1932
5.7,1/1/1933
8.7,1/1/1934
36.1,1/1/1935
79.7,1/1/1936
114.4,1/1/1937
109.6,1/1/1938
88.8,1/1/1939
67.8,1/1/1940
47.5,1/1/1941
30.6,1/1/1942
16.3,1/1/1943
9.6,1/1/1944
33.2,1/1/1945
92.6,1/1/1946
151.6,1/1/1947
136.3,1/1/1948
134.7,1/1/1949
83.9,1/1/1950
69.4,1/1/1951
31.5,1/1/1952
13.9,1/1/1953
4.4,1/1/1954
38,1/1/1955
141.7,1/1/1956
190.2,1/1/1957
184.8,1/1/1958
159,1/1/1959
112.3,1/1/1960
53.9,1/1/1961
37.5,1/1/1962
27.9,1/1/1963
10.2,1/1/1964
15.1,1/1/1965
47,1/1/1966
93.8,1/1/1967
105.9,1/1/1968
105.5,1/1/1969
104.5,1/1/1970
66.6,1/1/1971
68.9,1/1/1972
38,1/1/1973
34.5,1/1/1974
15.5,1/1/1975
12.6,1/1/1976
27.5,1/1/1977
92.5,1/1/1978
155.4,1/1/1979
32.27,1/1/1980
54.25,1/1/1981
59.65,1/1/1982
63.62,1/1/1983"

library(chron)
library(zoo)
z <- read.zoo(textConnection(Lines), header = TRUE, sep = ",",
	FUN = function(x) as.yearmon(chron(x)), index = 2)
acf(z)


On Mon, Jan 19, 2009 at 9:34 AM, Harsh <singhalblr at gmail.com> wrote:
> Very Sorry for the oversight.
>
> The dataset that I have used is:
>
> Sunspots,Datefield
> 9.5,1/1/1900
> 2.7,1/1/1901
> 5,1/1/1902
> 24.4,1/1/1903
> 42,1/1/1904
> 63.5,1/1/1905
> 53.8,1/1/1906
> 62,1/1/1907
> 48.5,1/1/1908
> 43.9,1/1/1909
> 18.6,1/1/1910
> 5.7,1/1/1911
> 3.6,1/1/1912
> 1.4,1/1/1913
> 9.6,1/1/1914
> 47.4,1/1/1915
> 57.1,1/1/1916
> 103.9,1/1/1917
> 80.6,1/1/1918
> 63.6,1/1/1919
> 37.6,1/1/1920
> 26.1,1/1/1921
> 14.2,1/1/1922
> 5.8,1/1/1923
> 16.7,1/1/1924
> 44.3,1/1/1925
> 63.9,1/1/1926
> 69,1/1/1927
> 77.8,1/1/1928
> 64.9,1/1/1929
> 35.7,1/1/1930
> 21.2,1/1/1931
> 11.1,1/1/1932
> 5.7,1/1/1933
> 8.7,1/1/1934
> 36.1,1/1/1935
> 79.7,1/1/1936
> 114.4,1/1/1937
> 109.6,1/1/1938
> 88.8,1/1/1939
> 67.8,1/1/1940
> 47.5,1/1/1941
> 30.6,1/1/1942
> 16.3,1/1/1943
> 9.6,1/1/1944
> 33.2,1/1/1945
> 92.6,1/1/1946
> 151.6,1/1/1947
> 136.3,1/1/1948
> 134.7,1/1/1949
> 83.9,1/1/1950
> 69.4,1/1/1951
> 31.5,1/1/1952
> 13.9,1/1/1953
> 4.4,1/1/1954
> 38,1/1/1955
> 141.7,1/1/1956
> 190.2,1/1/1957
> 184.8,1/1/1958
> 159,1/1/1959
> 112.3,1/1/1960
> 53.9,1/1/1961
> 37.5,1/1/1962
> 27.9,1/1/1963
> 10.2,1/1/1964
> 15.1,1/1/1965
> 47,1/1/1966
> 93.8,1/1/1967
> 105.9,1/1/1968
> 105.5,1/1/1969
> 104.5,1/1/1970
> 66.6,1/1/1971
> 68.9,1/1/1972
> 38,1/1/1973
> 34.5,1/1/1974
> 15.5,1/1/1975
> 12.6,1/1/1976
> 27.5,1/1/1977
> 92.5,1/1/1978
> 155.4,1/1/1979
> 32.27,1/1/1980
> 54.25,1/1/1981
> 59.65,1/1/1982
> 63.62,1/1/1983
>
> I convert the Datefield column to class Date, by using Datefield<-
> as.Date(Datefield,format="%m/%d/%Y")
> Then I create a time series object by using
> dtxts<- xts(Sunspots,order.by = Datefield)
>
> On using the ACF function
> acf(dtxts,plot= TRUE,xaxt = "n",col="red",na.action = na.omit)
> I get the error
> Error in na.omit.ts(as.ts(x)) : time series contains internal NAs
>
> The peculiar issue is that I do not get the above error on
> - R 2.7 on Ubuntu in India
> - R 2.8 on Windows in India
>
> The error appears when I execute this code on an R 2.8 implementation
> running on a Fedora machine in the USA.
>
> Is it a problem with the locale?
>
> Thanks for the help. Really appreciate it.
>
> Harsh Singhal
> Decisions Systems
> Mu Sigma Inc.
> Chicago, USA
>
>        [[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.
>




More information about the R-help mailing list