I am trying to determine the serial autocorrelation in my precipitation data. I have data for 30 years (1986-2015) and I am using the "ZYP" package in R and the Zhang method described in: Zhang, X., Vincent, L.A., Hogg,W.D. and Niitsoo, A., 2000. Temperature and Precipitation Trends
in Canada during the 20th Century. Atmosphere-Ocean 38(3): 395-429.
My code is simple
setwd("C:/Users/sch298/Documents/Kentucky River weather data/Daily homogenization")
y=read.csv("P-GHCNDUSC00150624.csv",as.is=TRUE)[, 2]
zyp.trend.vector(y, x=1:length(y), "zhang", T, T)
It ran fine but all the trend estimates are zero including the lower and upper bounds. I am sure something is wrong. I have attached my data file and output screenshot. Can anyone please help me to understand what might be wrong?
Thanks
Som
You can just simply use ACF function in R to detect correlation. The Zhang method may possibly been influenced by decomposition theory. For trend detection, i suggest you to use modified Mann-Kendall test for correlated data.
Before applying the prewhitening method you need to check first if the data is autocorrelated. if yes, now you can performed the analysis, if not only a man-kendall analysis is sufficient to detect a serial correlation.
- Badges
- Science topic
- Similar topics
- Filing