Hi guys,
I raised a question when conducting the Two-Sample Kolmogorov-Smirnov Test. So, I have two sets of data, and I'd like to compare their similarity in distribution. After I ran the K-S test, I got a P < 0.001, which reject the hypothesis and suggests two datasets are not similar in distribution (please see attached the figure). However, I also got a D = 0.121 from the result. As far as I know, when D close to 1, it indicates two datasets have different distributions, and when D close to 0, it indicates two datasets have similar distributions. When looking at this very low D value, does it mean the cumulative distributions between two datasets are actually similar? It seems controversial.
So, how should I interpret the result? And is it reasonable to use the D value to quantify the similarity between these two datasets?
Thanks in advance
Usted esta comparando dos grupos de datos en cuanto a normalidad , usando el esatdistico de K-s , kolmogorov smirnov, , primero deben ser muestras grandes, mayores que 30, vamos a las hipotesis:
ho: las observaciones son normales
h1: las observaciones no son normales
sin hablar de alfa ni de los otros pasos que faltan en el test, si ambos grupos de datos son normales me quedo con el mayor p-valor , siendo ambos mayores que alfa, es decir, ambos p-valores deben ser mayores que alfa, para no rechazar ho, y luego me quedo con el set de datos que posea mayor p-valor, por que !!!, bueno esto lleva otro analisis mas profundo !!!!, pero sencillo, se lo puedo explicar , digame usted si desea la explicación !!!
- Badges
- Science topic
- Similar topics
- Social Psychology
- Attitude