Generating data set timestamps based on distribution

I have a data set of 4600 points which are messages that another program I have made will analyse. To do this, each message is tagged with a timestamp in whole seconds of when it will be analysed, between 0 and 600. For this data set, I need the messages to be analysed during a specific time range, i.e. 240 - 440 seconds. The aim of this is to replicate social media posts about a topic, with a sudden explosion of posts followed by gently tail off. A trace of the posts would be something like the orange line below.

325px-Gamma_distribution_pdf.svg.png

So what I need help with is creating a function (defined or custom via VBA) that takes parameters of start and end points for the time range (i.e. 240, 440) and to generate a timestamp for each of the 4600 data points to distribute them in a way similar to the line on the graph.

Any help much appreciated!

Thanks,

Dan

Sorry, I don't follow your question. What would you like an example of?

Your post barely contains enough information for somebody to help who maybe works on similar tasks and already has an understanding of what you're doing.

For the rest of us it's like taking a high tech military plane to Joe's back street garage for an oil change, until he has enough information to do the job he's either going to be looking at it with a blank expression, or making a complete mess of things.

I see.

Right! My dataset is 4600 messages that replicate social media posts over a period of time. Each is to be read into a data analyser based on an associated timestamp in whole seconds between 0 and 600. For example, the analyser will call in all posts with timestamps between 5 and 9. It is these timestamps that I need to be able to generate to adjust when they will be processed. To do this I want to be able to provide a start and end time for the timestamps, 240 and 440 seconds for example. This gives a block of time where all 4600 messages will be called into the anlayser. The reason I need them to follow a distribution during that time range, such as in the graph above, is the nature of a sudden "explosion" of posts about a topic that then die off to nothing. Obviously, a lot of these posts will have the same timestamp. This means the frequency of posts being analysed matches one of the above distributions with a few hundred having the same timestamp to begin with, replicating the spike in talk about the topic, to a few or individual posts towards the end as talk quietens down.

So for the posts, a column of values for timestamp will be generated to something like:
Post1 - 240
Post2 - 240
Post3 - 241
...
Post4600 - 440

Hope that makes things a bit clearer

I have used the above formula over my dataset and then rounded down to give me integers. I have then counted the number of timestamps in each bin and plotted a histogram to view my distribution. It seems to give 36 values in each bin (between 240 and 449).

Am I doing something wrong? I was expecting the shape to be similar to the orange line in the diagram in my original post.

My spreadsheet is attached.

Thanks!

Sorry I forgot to respond sooner, this is exactly what I needed!