Here is one thought. Assign each month a value of 0, 1 or 2. Then do a
simple linear regression analysis where the value of the month
is the independent variable. You can also do multiple linear regression
with the value you assigned to the month plus the other factors that
you believe are causing a change to your data. Time is the one that
comes to my mind. You can do this with the standard R function lm.
I hope this helps.
Bob
On 1/30/2017 9:11 AM, Kwesi Quagraine wrote:
Hello, I have a data with two variables nodes and index, I want to extract
3 months seasons, with a shift of 1 month, that is, DJF, JFM, FMA etc to
OND. Was wondering how to go about it. Kindly find data sample below, data
is in csv format.
Any help will be appreciated.
My data sample;
era...1. Node_freq MEI
1 1980-01-01 -0.389855332 0.3394196488
2 1980-02-01 -0.728019153 0.2483738232
3 1980-03-01 -1.992457784 0.3516954904
4 1980-04-01 0.222760284 0.5736836269
5 1980-05-01 0.972601798 0.6289249144
6 1980-06-01 0.570725954 0.5736836269
7 1980-07-01 -0.977966324 0.4120517119
8 1980-08-01 0.056128836 -0.0104418383
9 1980-09-01 0.987304573 -0.0687520861
10 1980-10-01 1.188242495 -0.1403611624
11 1980-11-01 1.693037763 -0.0963727298
12 1980-12-01 1.173539720 -0.2539126977
13 1981-01-01 0.423698206 -0.6140040528
14 1981-02-01 -2.208098481 -0.5209122536
15 1981-03-01 -0.786830252 0.1133395650
16 1981-04-01 -0.110502611 0.3302127675
17 1981-05-01 -1.272021820 -0.1894645290
18 1981-06-01 0.394292656 -0.3736021538
19 1981-07-01 1.452892441 -0.4032687711
20 1981-08-01 0.698150002 -0.4441882433
21 1981-09-01 0.997106423 -0.1720737534
22 1981-10-01 0.247264908 -0.2436828296
23 1981-11-01 0.771663876 -0.3909929295
24 1981-12-01 -0.316341458 -0.4943145967
Regards,
Kwesi
______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
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.