I am not statistical faculty student. But, for climatic research, some time I need several statistical test. General I use various type of software like XLSTAT, SPSS, Matlab function or R Programming language to solve my statistical problem.
Now I want to write my own function to calculate several Homogeneity test like SNHT, Pettitt, Buishand Range or Von Neumann (Previously I Use XLSTAT to calculate Homogeneity test). Because of, I want to Batchly calculate Homogeneity for many station data.
The theory of those Homogeneity test are available in many research paper (Some paper listed in below). But, I can not find, how to calculate p value for those test. Can any one help me, to solve this problem ?
- Lin, Ng Jing, et al. "HOMOGENEITY ANALYSIS OF RAINFALL IN KELANTAN, MALAYSIA." Jurnal Teknologi 76.15 (2015).
- Taxak, Arun Kumar, A. R. Murumkar, and D. S. Arya. "Long term spatial and temporal rainfall trends and homogeneity analysis in Wainganga basin, Central India." Weather and Climate Extremes 4 (2014): 50-61.
There are some basic ideas in the link.
http://www.dummies.com/how-to/content/how-to-determine-a-pvalue-when-testing-a-null-hypo.html
We can see http://www.dummies.com/how-to/conte...value-when-testing-a-null-hypo.html
You can see http://www.dummies.com/how-to/conte...value-when-testing-a-null-hypo.html
I think there is a R command "ncvTest" test for non-constant variance (in package "car")
Si se entiende por falta de homogeneidad, la presencia de cambios graduales (tendencia) o abruptos (saltos), depende del test empleado. Para tendencia, usando un método paramétrico la instrucción en EXCEL es:
=DISTR.T(ABS(t calculada);(N-1);2); también se emplea en el método no paramétrico SROC (Sperman) y para verificar los saltos en el test t de Student Secuencial.
Para verificar la tendemncia en métodos no parametricos como: Mann y Kendall, Maann y Kendall con corrección de varianza o Mann y Kendall con preblanqueo se emplea:
=2*(1-DISTR.NORM.ESTAND(ABS(Z))).
Espero haber podido ayudar.
en EXCEL:
=DISTR.T(ABS(estadístico); número de datos - 2;2)
o
=2*(1-DISTR.NORM.ESTAND(ABS(estadistico)))
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS