Time shift with distribution

PDG

New Member
#1
Hello,

I'm looking for a formula that accounts for a timeshift that is distributed (weibull/lognormal)

For example.

I have "hits" that fall into a week.
W1: 1 1 1 1 1 1
Total hits W1=6
W2: 1 1 1 1
Total hits W2=4
W3: 1 1 1 1 1
Total hits W3=5
W4: 1 1 1 1 1 1 1
Total hits W4=7

Each week is ralated to a volume
W1 = 1000
W2 = 2000
W3 = 1500
W4 = 1700

The fraction would be Y(Wx) = Total hits / Volume

Now for each hit there is a time shift, lets say 1 week. And 1 week is the modus of the shift distribution and the distribution is a Weibull or lognormal.
How do I bring this correction with its distribution properly into the formula?

Thank you.
 

Miner

TS Contributor
#2
This sounds like you are looking for the threshold parameter (gamma) for the 3-parameter lognormal/weibul distributions. The threshold parameter defines the lower boundary for each distribution.
 

PDG

New Member
#3
3-par logn had a better fit in the probability plot.
The parameters are:
Lognormal; Loc=2,27719; Scale=0,859713; Thresh=0,293398