1-出口数据的平稳性分析

首先根据时序图判断

　

　　序列呈增长趋势和周期趋势，所以初步判定为非平稳序列

自相关图

　

进行ADF单位根检验

　　R语言adfTest函数中对序列三种假设进行检验

　　　　　　nc 无常数均值，无趋势类型

　　　　　　c   有常数均值，无趋势类型

　　　　　　ct  有常数均值，有趋势类型

library(readxl)
library(TSA)
library(fUnitRoots)
library(fBasics)
library(tseries)
# 导入数据
xlsx_1<-read_excel("~/Desktop/graduation design/WIND/export for USA.xlsx", 
                   sheet = "Sheet2")
# 提取数据
mon <- xlsx_1[,2]

# 构建时间序列
mon.timeseries <- ts(mon,start = c(1995,1),end = c(2019,12),frequency = 12)
# 画出时序图和自相关图
plot(mon.timeseries)
acf(mon.timeseries)

# 对时间序列进行DF检验
print(adfTest(mon.timeseries,lags = 1,type = "c"))
print(adfTest(mon.timeseries,lags = 1,type = "nc"))
print(adfTest(mon.timeseries,lags = 1,type = "ct"))



Title:
 Augmented Dickey-Fuller Test

Test Results:
  PARAMETER:
    Lag Order: 1
  STATISTIC:
    Dickey-Fuller: -1.5387
  P VALUE:
    0.4841 

Description:
 Mon Mar 30 21:47:39 2020 by user: 


Title:
 Augmented Dickey-Fuller Test

Test Results:
  PARAMETER:
    Lag Order: 1
  STATISTIC:
    Dickey-Fuller: -0.2097
  P VALUE:
    0.5495 

Description:
 Mon Mar 30 21:47:39 2020 by user: 


Title:
 Augmented Dickey-Fuller Test

Test Results:
  PARAMETER:
    Lag Order: 1
  STATISTIC:
    Dickey-Fuller: -6.2865
  P VALUE:
    0.01 

Description:
 Mon Mar 30 21:47:39 2020 by user: 

Warning message:
In adfTest(mon.timeseries, lags = 1, type = "ct") :
  p-value smaller than printed p-value

根据p值，可以判断接受原假设，即非平稳