R 处理时间序列数据中的连续缺失值_R_Time Series_Na

R 处理时间序列数据中的连续缺失值

R 处理时间序列数据中的连续缺失值,r,time-series,na,R,Time Series,Na,我有如下所示的时间序列数据 2015-04-26 23:00:00 5704.27388916015661380 2015-04-27 00:00:00 4470.30868326822928793 2015-04-27 01:00:00 4552.57241617838553793 2015-04-27 02:00:00 4570.22250032825650123 2015-04-27 03:00:00 NA 2015-04-27 04:00:00 NA 2015-04-27 0

我有如下所示的时间序列数据

2015-04-26 23:00:00  5704.27388916015661380
2015-04-27 00:00:00  4470.30868326822928793
2015-04-27 01:00:00  4552.57241617838553793
2015-04-27 02:00:00  4570.22250032825650123
2015-04-27 03:00:00  NA
2015-04-27 04:00:00  NA
2015-04-27 05:00:00  NA
2015-04-27 06:00:00 12697.37724086216439900
2015-04-27 07:00:00  5538.71119009653739340
2015-04-27 08:00:00    81.95060647328695325
2015-04-27 09:00:00  8550.65816895300667966
2015-04-27 10:00:00  2925.76573206583680076

我应该如何处理连续的NA值。在只有一个NA的情况下，我使用NA项极值的平均值。是否有处理连续缺失值的标准方法？

zoo软件包有几个处理

NA

值的函数。以下功能之一可能适合您的需要：

```
na.locf
```
：最后一次观察结转。使用参数
```
fromLast=TRUE
```
对应于下一次反向观测（NOCB）
```
na.aggregate
```
：用一些聚合值替换
```
na
```
。默认聚合函数是
```
平均值
```
，但您也可以指定其他函数。有关更多信息，请参见
```
？na.aggregate
```
```
na.近似值
```
：
```
na
```
替换为线性插值

您可以比较结果以了解这些功能的作用：

library(zoo)
df$V.loc <- na.locf(df$V2)
df$V.agg <- na.aggregate(df$V2)
df$V.app <- na.approx(df$V2)

使用数据：

df <- structure(list(V1 = structure(c(1430082000, 1430085600, 1430089200, 1430092800, 1430096400, 1430100000, 1430103600, 1430107200, 1430110800, 1430114400, 1430118000, 1430121600), class = c("POSIXct", "POSIXt"), tzone = ""), V2 = c(5704.27388916016, 4470.30868326823, 4552.57241617839, 4570.22250032826, NA, NA, NA, 12697.3772408622, 5538.71119009654, 81.950606473287, 8550.65816895301, 2925.76573206584)), .Names = c("V1", "V2"), row.names = c(NA, -12L), class = "data.frame")

或

df <- structure(list(V1 = structure(c(1430082000, 1430085600, 1430089200, 1430092800, 1430096400, 1430100000, 1430103600, 1430107200, 1430110800, 1430114400, 1430118000, 1430121600), class = c("POSIXct", "POSIXt"), tzone = ""), V2 = c(5704.27388916016, 4470.30868326823, 4552.57241617839, 4570.22250032826, NA, NA, NA, 12697.3772408622, 5538.71119009654, 81.950606473287, 8550.65816895301, 2925.76573206584)), .Names = c("V1", "V2"), row.names = c(NA, -12L), class = "data.frame")

 library("imputeTS")

 # Moving Average Imputation
 na.ma(df$V2)

 # Imputation via Kalman Smoothing on structural time series models 
 na.kalman(df$V2)

 # Just interpolation but with some nice options (linear, spline,stine)
 na.interpolation(df$V2)

library("forecast")

#Interpolation via seasonal decomposition and interpolation
na.interp(df$V2)