I would like to know the value of the upper adjacent value in a boxplot I have, so that I can exclude all the outliers and compare the regressional results. However, I cannot find an option to include the value. The help sheet gives following formula
tells me that my x 75 is 160000 and my x25 is 25000
so my U is = 295000
However, when I graph the variable this doesn't seem very realistic. I attached the graph. But the upper value shoud be around 360000.
Am I doing something wrong with the math, or what am I missing? How can I figure the value out? Or is there an easier way to find out the outliers? Note that this is just a small sample used in the graph, I have a lot of outliers so playing with the graph settings till I can physically see the first outlier (and exclude all observations with values that are similar to it or above) really isn't something I want to do.
Let x represent a variable for which adjacent values are being calculated. Define x(i) as the ith ordered value of x, and define x[25] and x[75] as the 25th and 75th percentiles. graph box — Box plots 17 Define U as x[75] + 3/ 2 (x[75] − x[25]). The upper adjacent value is defined as xi , such that x(i) ≤ U and x(i+1) > U.
Define L as x[25] − 3 /2 (x[75] − x[25]). The lower adjacent value is defined as xi , such that x(i) ≥ L and x(i−1) < L.
Define L as x[25] − 3 /2 (x[75] − x[25]). The lower adjacent value is defined as xi , such that x(i) ≥ L and x(i−1) < L.
Code:
su var, d
so my U is = 295000
However, when I graph the variable this doesn't seem very realistic. I attached the graph. But the upper value shoud be around 360000.
Am I doing something wrong with the math, or what am I missing? How can I figure the value out? Or is there an easier way to find out the outliers? Note that this is just a small sample used in the graph, I have a lot of outliers so playing with the graph settings till I can physically see the first outlier (and exclude all observations with values that are similar to it or above) really isn't something I want to do.
Comment