我正在进行时间序列分析。我想分析1971年至2015年的季度生产率数据。然而,生产率用2010年= 1的生产力指数表示。
这意味着2010年4季度的平均值等于1.这被认为是初始值。然后,相对于该值表示生产率的增加或减少。
> dput(head(prod_ts,179))
structure(c(0.4652, 0.4721, 0.4808, 0.4827, 0.4814, 0.493, 0.4936,
0.5002, 0.5221, 0.5228, 0.5193, 0.518, 0.5058, 0.5152, 0.5193,
0.5132, 0.5163, 0.5089, 0.5088, 0.517, 0.5269, 0.5229, 0.5279,
0.5384, 0.5393, 0.5369, 0.5409, 0.5482, 0.5498, 0.5543, 0.5594,
0.561, 0.5553, 0.5782, 0.5631, 0.5679, 0.5632, 0.5545, 0.5565,
0.5545, 0.5552, 0.5599, 0.5707, 0.5742, 0.5787, 0.5884, 0.5916,
0.5984, 0.6102, 0.6152, 0.6185, 0.6214, 0.6244, 0.6173, 0.6182,
0.6247, 0.6304, 0.639, 0.6377, 0.6412, 0.6504, 0.6584, 0.6633,
0.6736, 0.6753, 0.6815, 0.6925, 0.6937, 0.6995, 0.6978, 0.7034,
0.7037, 0.7013, 0.6999, 0.6977, 0.6982, 0.7059, 0.7105, 0.7021,
0.6992, 0.7016, 0.7051, 0.7098, 0.7187, 0.7285, 0.7436, 0.7518,
0.7638, 0.7707, 0.7748, 0.7816, 0.7884, 0.7935, 0.8039, 0.8123,
0.818, 0.8232, 0.823, 0.8284, 0.8266, 0.8345, 0.8359, 0.8399,
0.8434, 0.8496, 0.8545, 0.8577, 0.8661, 0.8696, 0.8734, 0.8759,
0.881, 0.8837, 0.8824, 0.8938, 0.903, 0.9101, 0.9131, 0.9129,
0.915, 0.9218, 0.9266, 0.9323, 0.9324, 0.9371, 0.9392, 0.9467,
0.9497, 0.956, 0.9603, 0.9689, 0.9747, 0.9744, 0.9782, 0.9788,
0.977, 0.9814, 0.9916, 0.9977, 1.0141, 1.0108, 1.0109, 1.009,
1.0116, 1.0214, 1.0243, 1.0276, 1.0304, 1.0295, 1.0212, 1.0097,
0.9894, 0.979, 0.9861, 0.9883, 0.9898, 0.9982, 1.001, 0.999,
1.0018, 1.0037, 1.005, 1.0146, 1.0144, 1.0168, 1.0088, 1.0171,
1.0106, 1.0193, 1.0214, 1.0244, 1.0235, 1.0239, 1.0269, 1.0305,
1.0347, 1.0309, 1.0383, 1.0358), .Tsp = c(1971, 2015.5, 4), class = "ts")
这对我的分析来说不是很有用。我希望表达1971年的季度总和,使其总和为1.然后,时间序列的所有剩余值(1972年至2015年的每个季度)表示为具有初始值的时间序列( 1971,1)。
答案 0 :(得分:0)
如果我理解正确你想要平均1971年的值(Q1到Q4)为1,就像现在的2010年一样。如果是这样,这应该解决它(假设你的数据在Function Function_Name(x, y)
sumx = 0
For i = x To x
If y = "text1" Then sumx = x * 12
If y = "text2" Then sumx = x * 6
sumx = sumx + sumx
Next i
Function_Name = sumx
End Function
:
df