2

I have a set of observed data and created an empirical cumulative distribution using Excel. I want to use this CDF to find probabilities like P(x < X) or P (X1 < x < X2 ).

The way I created the CDF is to arrange the data in ascending order and then create a column next to it with the probabilities:

https://i.stack.imgur.com/VsXhD.png

I have 4,121 records and the sample here is for four records. Once I have this calculation done, the curve is plotted using xy scatter plot for Data in the x-axis and Probability in the y-axis. This is how I created the CDF.

How can I find probability below 2.5, P(x<=2.5), or P( 970 < x < 980 )?

I hope there is an easy way because I will have hundreds of probabilities to find.

pnuts
  • 58,317
  • 11
  • 87
  • 139
exlover
  • 69
  • 1
  • 2
  • 8
  • I am not too familiar with probabilities, but are you looking for something like the countif function? For example, =COUNTIF(A2:A500, "<2.5")/(499-COUNTBLANK(A2:A500) – Yaegz Oct 09 '15 at 20:48
  • No I am looking for something to lookup for a value closest to the 2.5 for instance and then maybe interpolate for the probability. In the picuture for instance, I know that 2.08 has a probability of 0.000727979 and 2.71 has a probability of 0.000970638.. Therefore, the probability of 2.5 must be somewhere between 0.000727979 and 0.000970638. I know how to calculate this by hand, but if I have a huge set of probabilities that I need to find then this method is time consuming. – exlover Oct 09 '15 at 21:02
  • 1
    Can you tell us how you are doing it by hand? – Simon White Oct 19 '15 at 08:13

0 Answers0