R：不完全なtimeseriesを行列に変換する

Question

私は月ごとの時系列を持ち、それを行列に変換する必要があります。 （6年、最後の4ヶ月は欠落しています）

 Jan  Feb  Mar  Apr  May  Jun  Jul  Aug  Sep  Oct  Nov  Dec 
1 62.17416 66.34748 69.58154 64.92033 70.51246 70.83153 74.19475 72.29366 77.97062 71.37903 67.08182 69.10599 
2 59.41197 61.44036 68.97711 72.43558 69.92715 74.20776 79.19150 71.48650 77.56661 67.44390 67.05218 66.43690 
3 58.79278 63.26696 70.97837 71.79423 72.04037 71.61184 72.64526 70.38479 72.43492 64.42046 66.87742 66.31345 
4 58.06906 60.35554 70.91920 65.76937 68.85464 68.77921 70.09373 66.03685 71.38322 66.42558 67.62292 70.25223 
5 60.59774 64.66411 73.27853 73.11031 73.57735 73.77766 78.45258 74.64698 79.07911 75.84108 75.82478 79.42484 
6 69.30907 70.80211 78.38037 77.63537 79.67396 80.15397 80.45571 76.67389 

時系列の最後の年には値がありません。 行列に変換するとき、関数は時系列の最初の値を使用して欠損値を置き換えます。

> matrix(xx,nrow=6,ncol=12,byrow = TRUE) 
     [,1]  [,2]  [,3]  [,4]  [,5]  [,6]  [,7]  [,8]  [,9] [,10] [,11] [,12] 
[1,] 62.17416 66.34748 69.58154 64.92033 70.51246 70.83153 74.19475 72.29366 77.97062 71.37903 67.08182 69.10599 
[2,] 59.41197 61.44036 68.97711 72.43558 69.92715 74.20776 79.19150 71.48650 77.56661 67.44390 67.05218 66.43690 
[3,] 58.79278 63.26696 70.97837 71.79423 72.04037 71.61184 72.64526 70.38479 72.43492 64.42046 66.87742 66.31345 
[4,] 58.06906 60.35554 70.91920 65.76937 68.85464 68.77921 70.09373 66.03685 71.38322 66.42558 67.62292 70.25223 
[5,] 60.59774 64.66411 73.27853 73.11031 73.57735 73.77766 78.45258 74.64698 79.07911 75.84108 75.82478 79.42484 
[6,] 69.30907 70.80211 78.38037 77.63537 79.67396 80.15397 80.45571 76.67389 62.17416 66.34748 69.58154 64.92033 

これを回避する方法、または最後の値を "NA"で置き換える方法はありますか？

Answer 1

xxは、最後の注記に記載されている時系列です。次のように私たちが知らなくても行列に変換することができます次に

1）はどのように多くの不足している項目がある：

  1  2  3  4  5  6  7  8 
2011 62.17416 66.34748 69.58154 64.92033 70.51246 70.83153 74.19475 72.29366 
2012 59.41197 61.44036 68.97711 72.43558 69.92715 74.20776 79.19150 71.48650 
2013 58.79278 63.26696 70.97837 71.79423 72.04037 71.61184 72.64526 70.38479 
2014 58.06906 60.35554 70.91920 65.76937 68.85464 68.77921 70.09373 66.03685 
2015 60.59774 64.66411 73.27853 73.11031 73.57735 73.77766 78.45258 74.64698 
2016 69.30907 70.80211 78.38037 77.63537 79.67396 80.15397 80.45571 76.67389 
      9  10  11  12 
2011 77.97062 71.37903 67.08182 69.10599 
2012 77.56661 67.44390 67.05218 66.43690 
2013 72.43492 64.42046 66.87742 66.31345 
2014 71.38322 66.42558 67.62292 70.25223 
2015 79.07911 75.84108 75.82478 79.42484 
2016  NA  NA  NA  NA 

2：

tt <- time(xx) 
tapply(xx, list(floor(tt), cycle(tt)), c) 

は、この6×12のマトリックスを与えます）もう1つの可能性は、長さを12の次の最も高い倍数に強制して、行列を正確に塗りつぶし、それを再形成することです。これは、問題の場合のように、追加されるNAsが最後にある場合にのみ機能します。

m <- xx # copy xx in case we still need it later 
freq <- frequency(xx) # 12 
length(m) <- freq * ceiling(length(m)/freq) # pads with NAs to achieve length 
m <- matrix(m,, freq, byrow = TRUE) # reshape m into a matrix 

与える：

> m 
     [,1]  [,2]  [,3]  [,4]  [,5]  [,6]  [,7]  [,8] 
[1,] 62.17416 66.34748 69.58154 64.92033 70.51246 70.83153 74.19475 72.29366 
[2,] 59.41197 61.44036 68.97711 72.43558 69.92715 74.20776 79.19150 71.48650 
[3,] 58.79278 63.26696 70.97837 71.79423 72.04037 71.61184 72.64526 70.38479 
[4,] 58.06906 60.35554 70.91920 65.76937 68.85464 68.77921 70.09373 66.03685 
[5,] 60.59774 64.66411 73.27853 73.11031 73.57735 73.77766 78.45258 74.64698 
[6,] 69.30907 70.80211 78.38037 77.63537 79.67396 80.15397 80.45571 76.67389 
     [,9] [,10] [,11] [,12] 
[1,] 77.97062 71.37903 67.08182 69.10599 
[2,] 77.56661 67.44390 67.05218 66.43690 
[3,] 72.43492 64.42046 66.87742 66.31345 
[4,] 71.38322 66.42558 67.62292 70.25223 
[5,] 79.07911 75.84108 75.82478 79.42484 
[6,]  NA  NA  NA  NA 

注：再現形態における入力時系列xxである：

xx <- ts(c(62.17416, 66.34748, 69.58154, 64.92033, 70.51246, 
70.83153, 74.19475, 72.29366, 77.97062, 71.37903, 67.08182, 69.10599, 
59.41197, 61.44036, 68.97711, 72.43558, 69.92715, 74.20776, 79.1915, 
71.4865, 77.56661, 67.4439, 67.05218, 66.4369, 58.79278, 63.26696, 
70.97837, 71.79423, 72.04037, 71.61184, 72.64526, 70.38479, 72.43492, 
64.42046, 66.87742, 66.31345, 58.06906, 60.35554, 70.9192, 65.76937, 
68.85464, 68.77921, 70.09373, 66.03685, 71.38322, 66.42558, 67.62292, 
70.25223, 60.59774, 64.66411, 73.27853, 73.11031, 73.57735, 73.77766, 
78.45258, 74.64698, 79.07911, 75.84108, 75.82478, 79.42484, 69.30907, 
70.80211, 78.38037, 77.63537, 79.67396, 80.15397, 80.45571, 76.67389 
), start = c(2011, 1), freq = 12)