多列时间序列的线性回归 [英] Linear regression of time series over multiple columns

问题描述

我有以下问题.我想以矩阵形式计算年度时间序列的回归.总共，我从要检查的区域的网格点提取了56个时间序列，这样我每年就有56个值.我已将所有值绘制为图中的点.现在，我想在该图中添加一条回归线，其中包含所有数据.

I have the following problem. I want to compute the regression of an annual time series in matrix form. In total, I have 56 time series I extracted from gridpoints of an area I want to examine, so that I've got 56 values per year. I've plotted all values as points in a figure. Now I want to add a regression line to this figure, which contains all data.

我的目标是计算整个矩阵的回归.

My goal is to compute the regression for the whole matrix.

library(zoo)

pdf(file="/home/user/name.pdf", pointsize=20, onefile = FALSE, width=18, height=11, paper = "special")
plot(mat.zoo[,1], pch=20, type="p", ylim=c(8,max(mat.zoo)),
     yaxt = "n", xaxt = "n", lwd = 1.5, main = "Some title", 
     ylab = "ylabtext", xlab ="", col = "black")

tt <- time(mat.zoo)
ix <- seq(1, length(tt), by=1) #every year a tick
labs <- format(tt[ix])
axis(side = 1, at = tt[ix], labels = labs,  tcl = -0.7, cex.axis = 1)

for (i in 2:ncol(mat.zoo)) {  
  #plot every column
  points(mat.zoo[,i], pch=20, lwd = 1.5)
}

## create ticks at every first y value 
axis(side = 2, at = seq(0, max(mat.zoo), by = 1), labels = FALSE)
iy <- seq(0, max(mat.zoo), 2)
axis(side = 2, at = iy, cex.axis = 1)

#this line doesn't work
abline( lm(mat.zoo ~ tt), col="light blue", lwd=3 )

dev.off()

图:

http://i.imgur.com/Ny5ERj1.png

一些示例数据，如果我使用dput()

Some sample data, if I use dput()

structure(c(14.6108611110572, 15.0943707315979, 16.4246753285039, 
15.4777258564571, 15.3910647660091, 14.9576052728563, 14.577379912167, 
15.6818364395762, 15.3935454316438, 14.6986382632628, 14.9616178291156, 
14.6208764396762, 17.4073263088521, 16.3932907105236, 16.4711871055354, 
15.7165524844793, 15.910687798697, 15.2800531253961, 16.2585353059321, 
14.9642915613775, 15.5682258772038, 15.7581733353644, 16.3600126905042, 
15.9906231843285, 16.4740591781654, 16.6207709477207, 16.7107736486755, 
15.3495937400046, 15.081738134456, 17.8213361743775, 17.0073514277019, 
16.0639354869614, 15.7564229038361, 16.4711872385234, 16.1474456418556, 
16.1012429675788, 14.935862417968, 14.649232718741, 14.7248073786802, 
16.3713171174875, 16.5047383689279, 15.6553509485205, 15.8069612127912, 
15.0880755914505, 15.9605131388024, 15.1647608142339, 15.0206531342878, 
15.8533914806642, 16.1936611693424, 14.4341552680467, 15.0030002589802, 
16.2373036559464, 15.4563912060316, 15.7540478676699, 14.4544119112367, 
14.1481450642128, 15.8808048538232, 15.3109864936677, 14.6184823877101, 
14.759740997088, 14.4554473653311, 17.1869089559961, 16.032779242263, 
15.9154018617995, 15.7003191635601, 16.1782858717824, 15.005330870126, 
16.1074524252519, 15.387333324397, 15.4238444378858, 15.7384875972114, 
16.3306448173221, 15.8050630623362, 16.5357139417134, 15.7318155157117, 
16.6027108391727, 15.3521994865507, 14.6028494060288, 17.0695642066462, 
16.5601941440799, 16.0704699986853, 15.9527367313925, 15.8492898967367, 
15.8094909404139, 15.9223122951851, 14.7427484210632, 14.3087395573591, 
14.9164340340289, 16.5109060631933, 16.1756705822203, 15.6869363317253, 
15.302941446409, 14.7871569748782, 16.2405108282472, 14.9030204259848, 
15.1076128392841, 15.7835364136346, 16.2406871099921, 13.9434587358454, 
14.8761562136977, 16.5604955686145, 15.3055531556642, 15.528200122034, 
14.3683664247369, 14.8660671257497, 16.2483828855783, 15.5912163679296, 
14.5206758668367, 15.0572249827849, 14.2126710362867, 16.7430589790551, 
15.913830135814, 15.5309377608968, 15.4301657033962, 16.1024796689616, 
14.9412190564665, 15.5415580911515, 15.6185795702858, 15.246965832492, 
15.5331896889331, 16.0527261022428, 16.2496153707101, 16.1013003488606, 
15.4012992267683, 16.6433171425044, 15.3443805149379, 15.0832591147848, 
17.2409394600713, 16.3670395392329, 15.8028463074112, 16.230362038712, 
15.8533914346074, 16.0962730847646, 15.4780493166121, 14.7644838005869, 
14.0160611132642, 14.7363498686371, 16.5339052116905, 16.1142787861115, 
15.1343982378726, 15.0479243093561, 14.8394739356758, 16.2015436792666, 
14.8852279610404, 15.171354759099, 15.8823805835669, 16.082598536468, 
13.6882801770178, 15.1822273858009, 16.7314060285488, 15.1822255101789, 
15.6470428935629, 14.6219009419668, 14.5344414346855, 17.0856674074961, 
15.6276761713817, 14.9656277726849, 15.0416098763217, 14.8660691394921, 
16.8350823196938, 15.7276830387531, 15.6464050524098, 15.7889210440969, 
15.8260661780512, 15.0685110014866, 15.5003231376182, 15.1818971179834, 
15.2523764253926, 15.2397513974873, 16.4076206985996, 16.2609962527472, 
15.9563455712026, 14.6758308266033, 15.9928106586864, 15.3388404382473, 
15.3352069271315, 17.491711796634, 16.3110401122382, 15.6722694212894, 
16.0979740581832, 15.9314161173117, 15.5309368794019, 15.5425227514293, 
14.8653903137068, 14.3680198631293, 14.6030824713595, 16.7764724794758, 
15.7590262768357, 14.9562687841841, 14.7258278360439, 15.00733114536, 
16.1086825085102, 14.8246425174662, 15.6697167018262, 15.5235314139726, 
15.810753562246), .Dim = c(49L, 4L), .Dimnames = list(c("1", 
"367", "732", "1097", "1462", "1828", "2193", "2558", "2923", 
"3289", "3654", "4019", "4384", "4750", "5115", "5480", "5845", 
"6211", "6576", "6941", "7306", "7672", "8037", "8402", "8767", 
"9133", "9498", "9863", "10228", "10594", "10959", "11324", "11689", 
"12055", "12420", "12785", "13150", "13516", "13881", "14246", 
"14611", "14977", "15342", "15707", "16072", "16438", "16803", 
"17168", "17533"), c("GB.1", "GB.2", "GB.3", "GB.4")), index = c(1960, 
1961, 1962, 1963, 1964, 1965, 1966, 1967, 1968, 1969, 1970, 1971, 
1972, 1973, 1974, 1975, 1976, 1977, 1978, 1979, 1980, 1981, 1982, 
1983, 1984, 1985, 1986, 1987, 1988, 1989, 1990, 1991, 1992, 1993, 
1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 
2005, 2006, 2007, 2008), class = "zoo")

推荐答案

尝试将您的lm语句修改为

lm(as.vector(mat.zoo) ~ rep(tt, length.out = length(mat.zoo)))

这篇关于多列时间序列的线性回归的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！